Canterbury Christ Church University’s repository of research outputs

http://create.canterbury.ac.uk

Copyright © and Moral Rights for this thesis are retained by the author and/or other copyright owners. A copy can be downloaded for personal non-commercial research or study, without prior permission or charge. This thesis cannot be reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder/s. The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the copyright holders.

When referring to this work, full bibliographic details including the author, title, awarding institution and date of the thesis must be given e.g. Danabalan, R. (2010) Mosquitoes of southern England and northern Wales: identification, ecology and host selection. Ph.D. thesis, University of Kent.

Contact: [email protected] Renita Danabalan

PhD Ecology

Mosquitoes of southern England and northern Wales: Identification, Ecology and Host selection.

Table of Contents:

Acknowledgements pages 1

Abstract pages 2

Chapter1: General Introduction Pages 3-26

1.1 History of Mosquito Systematics pages 4-11 1.1.1 Internal Systematics of the Subfamily Anophelinae pages 7-8 1.1.2 Internal Systematics of the Subfamily Culicinae pages 8-11 1.2 British Mosquitoes pages 12-20 1.2.1 Species List and Feeding Preferences pages 12-13 1.2.2 Distribution of British Mosquitoes pages 14-15 1.2.2.1 Distribution of the subfamily Culicinae in UK pages 14 1.2.2.2. Distribution of the genus Anopheles in UK pages 15 1.2.3 British Mosquito Species Complexes pages 15-20 1.2.3.1 The Anopheles maculipennis Species Complex pages 15-19 1.2.3.2 The Culex pipiens Species Complex pages 19-20 1.3 Mosquitoes as vectors of disease pages 20-21 1.4 Mosquito-borne diseases in the UK pages 21-25 1.4.1 Avian Malaria pages 21-22 1.4.2 West Nile virus (WNv) pages 22-23 1.4.3 Tahyna and Inkoo viruses pages 23-24 1.4.4 Sindbis virus pages 24 1.4.5 Arbo disease transmission in the UK pages 24-25 1.5 Overall project aims pages 25-26

Chapter 2: Field Collection of British Mosquitoes Pages 27-44

2.1 Introduction pages 28-33 2.1.1 British mosquitoes pages 28-32 2.1.2 Ecology of British mosquitoes pages 32-33 2.2 Materials and Methods pages 33-44 2.2.1 Collection of Immature stages pages 40-41 2.2.1.1 Individual rearing pages 40 2.2.1.2 Preparation of link-reared adults pages 41 2.2.2 Field collections of adult mosquitoes pages 41-43 2.2.3 Mosquito identification pages 44

Chapter 3: Molecular differentiation of the Pages 45-68 Maculipennis Group in the UK

3.1 Introduction pages 46-53 3.1.1 The taxonomic status of the Palaearctic pages 46-47 Maculipennis Group 3.1.2 Molecular differentiation of species in the pages 47-48 Maculipennis Group 3.1.3 Maculipennis Group and malaria transmission pages 48

ii 3.1.4 The Maculipennis Group in the UK pages 48-53 3.1.4.1 Distribution and ecology of the Maculipennis pages 49-50 Group in the UK 3.1.4.2 Role of the Maculipennis Group in pages 51-53 malaria transmission in the UK 3.2 Aims pages 54 3.3 Materials and Methods pages 54-59 3.3.1 Collection and identification of field-caught specimens pages 54-56 3.3.2 DNA extraction and ITS-PCR amplification pages 56 3.3.3 Direct sequencing of the ITS2 amplicon pages 56-58 3.3.4 Development of ITS PCR-RFLP assay pages 58-59 3.4 Results pages 60-64 3.4.1 ITS2 sequences pages 60-61 3.4.2 PCR-RFLP assay pages 62-63 3.4.3 Ecology pages 64 3.5 Discussion pages 64-68

Chapter 4: A critical assessment of molecular identification Pages 69-90 tools for the Palaearctic members of the Pipiens Group (Diptera: Culicidae)

4.1 Introduction pages 70-77 4.1.1 Taxonomic status and distribution of the Pipiens Group pages 70 4.1.2 The Pipiens Complex pages 71-73 4.1.3 Differentiation of Culex pipiens f. pipiens, pages 74-75 Cx. pipiens f. molestus and Cx. torrentium 4.1.4 Vector status of Cx. pipiens s.l. pages 76-77 4.2 Aims pages 77 4.3 Materials and Methods pages 77-82 4.3.1 Collection of Culex pages 77 4.3.2 Molecular identification pages 78-82 4.3.2.1 CQII Microsatellite Assay pages 78-80 4.3.2.2 Amplification of mtDNA COI gene fragment pages 81-82

4.4 Results pages 82-86 4.4.1 Collection of Culex mosquitoes pages 82 4.4.2 CQII Microsatellite assay pages 82-83 4.4.3 Congruence of CQII assay and mtDNA COI sequences pages 84-86 4.5 Discussion pages 87-90

Chapter 5: Host selection of British Mosquitoes Pages 91-119

5.1 Introduction pages 92-97 5.1.1 Mosquito-borne diseases in Europe pages 92 5.1.2 Importance of host selection pages 92-93 5.1.3 Host identification pages 93-94 5.1.4 British mosquitoes pages 94-97 5.2 Aims pages 97 5.3 Materials and Methods pages 97-105 5.3.1 Collection and identification of blood-fed females pages 97-100

iii 5.3.2 DNA extraction protocols pages 101 5.3.2.1 DNA extraction from host pages 101 serum and dried blood 5.3.2.2 DNA extractions for mosquito bloodmeal analysis pages 101 5.3.3 PCR amplification protocols pages 102-105 5.3.3.1 Universal Cytochrome B Oxidase (CytB) pages 102-103 amplification 5.3.3.2 Host-specific primer design and PCR optimisation pages 104-105 5.3.3.3 Determination of host selection pages 105 5.4 Results pages 106-113 5.4.1 Optimisation of Universal CytB primers pages 106-107 5.4.2 Optimisation of host specific primers pages 108-110 5.4.3 Host selection pages 111-113 5.5 Discussion pages 114-119

Chapter 6: Occurrence and Habitat Preference Pages 120-143 of mosquitoes in southern England and Wales

6.1 Introduction pages 121 6.1.1 Current occurence of British mosquitoes pages 121 6.2 Aims pages 121 6.3 Materials and Methods pages 122 6.3.1 Collection of larvae and pupae pages 122 6.3.2 Collection of resting and host seeking adult mosquitoes pages 122 6.4 Results pages 122-137 6.4.1 Characterisation of aquatic stages pages 124-127 6.4.2 Collected British mosquitoes detailed by genera pages 128-138 6.4.2.1 Genus Anopheles pages 128-131 6.4.2.1.1 Anopheles algeriensis pages 128 6.4.2.1.2 Anopheles atroparvus pages 128-129 6.4.2.1.3 Anopheles claviger pages 130 6.4.2.1.4 Anopheles daciae pages 130-131 6.4.2.1.5 Anopheles messeae pages 131 6.4.2.2 Genus Culex pages 132-134 6.4.2.2.1 Culex pipiens pages132 6.4.2.2.2 Culex torrentium pages 132-134 6.4.2.3 Genus Culiseta pages 135-136 6.4.2.3.1 Culiseta annulata pages 135 6.4.2.3.2 Culiseta subochrea pages 135-136 6.4.2.4 Genus Coquillettidia pages 137 6.4.2.5 Genus Dahliana pages 137 6.4.2.6 Genus Ochlerotatus pages 137-138 6.5 Discussion pages 138-143

Chapter 7: General Discussion Pages 144-151

7.1 Discussion pages 145-149 7.2 Future work pages 149-150

Bibliography pages 151-181

iv Total number of words: 47, 200.

v List of figures

Figure 1.1 Classification of the Family Culicidae after Edwards (1932)

Figure 1.2 Organogram of Family Culicidae after Belkin (1962) and Harbach & Kitching (1998).

Figure 1.3 Organogram of the subgenus Culex, following the most recent classification of Harbach (1988).

Figure 1.4 Distribution of Culicine mosquitoes in UK. (a) Distribution of genus Culex (modified from Rees & Snow et al., 1992; Snow et al., 1997); (b) Distribution of genus Culiseta and Orthopodomyia (modified from Rees & Snow, 1994; Snow et al., 1997) and (c) Distribution of the Aedes, Coquillettidia, Dahliana and Ochlerotatus genera (modified from Rees & Snow, 1994; Rees & Snow, 1995; Rees & Snow, 1996; Snow et al., 1997).

Figure 1.5 Distribution of Anopheles mosquitoes in UK. (a) Distribution of An. atroparvus, An. daciae and An. messeae (modified from Rees & Snow, 1990; Snow, 1998; Linton et al., 2002; Linton et al., 2005), and (b) Distribution of An. algeriensis, An. claviger and An. plumbeus (modified from Snow, 1990; Snow, 1998).

Figure 2.1 Collection form.

Figure 2.2 The reverse side of the field collection forms (see Figure 5.1) showing an example of the detailed notes kept of each individual larval rearing.

Figure 2.3 Collection of aquatic stages in natural habitats (a, b & d) streams and (e) a pond and artificial habitats including (c) disused bath tub, (f) moored boat and (g) fire buckets. Methods of collection for aquatic stages are shown in 2.3d.

Figure 2.4 Adult collection methods and resting habitats: (a) aspiration of adults resting on walls of a Brick Shelter in Kent, (b) collection cups used to transport collected resting adults, (c) Mosquito Magnet trap ® uses carbon dioxide, moisture and heat (bi-products of propane) and a chemical lure (1-Octen-3-ol) to attract host-seeking females (d) a brick shelter in Exminster Nature Reserve, Devon and (e) goat stable in Pettits Animal Farm, Norfolk.

Figure 3.1 Distribution of benign tertiary malaria in England from 1840 to 1910 (Kuhn et al., 2003). Intensity of colour signifies the annual number of cases per 100,000 inhabitants by county: Maroon =96-114, bright red =51-65, salmon pink =36-50 and light pink =9-20.

Figure 3.2 Species-diagnostic ITS2 products from the three British Maculipennis Group species following digestion with BstU I enzyme (CG↓CG). Lane 1: Hyperladder I 100 bp ladder (BioLine). Lanes 2 & 3: An.

vi atroparvus (445 bp, 42 bp). Lanes 4 & 5: An. messeae (332bp, 109bp, 42bp). Lanes 6 & 7: An. daciae (332bp, 59bp, 52bp, 42bp). Note: fragments under 100bp are generally not visible on this agarose gel.

Figure 3.3 The 489bp alignment of nuclear ITS2 sequences of An. messeae (485bp), An. daciae (485bp) and An. atroparvus (487bp). Enzyme BstU I (CG↓CG) cutting sites are highlighted in yellow. One site is present at 42 bp for all three species, a second site is present for both An. messeae and An. daciae at 378bp and a third site is present for An. daciae only at 437bp, thus the following species-diagnostic fragments are generated: An. atroparvus (445 & 42 bp), An. daciae (332, 59, 52 & 42bp) and An. messeae (332, 109 & 42bp).

Figure 4.1 Alignment of Cx. pipiens f. pipens, Cx. pipens f. molestus and Cx. torrentium using Mesquite (v2.6) showing the three restriction sites in the 710-bp fragment of the COI gene used by Shaikevich (2007). Figure 4.1(a) shows a single A-G polymorphism at base 205 that differentiates Cx. pipiens f. pipiens from Cx. pipiens f. molestus and Cx. torrentium. This difference is exploited in the first RFLP assay using enzyme Hae III (GG↓CC), which cuts the Cx. pipiens f. pipiens fragment into two, but leaves the Cx. pipiens f. molestus and Cx. torrentium uncut. Figures 4.1 (b & c) show the restriction sites of the second enzyme, Bc II (T↓GATCA), which cleaves Cx. torrentium at only one site (79bp) (Figure 3.3b) while digesting Cx. pipiens f. pipiens and Cx. pipiens f. molestus at two sites (79 and 485bp) (Figure 4.1.b, c).

Figure 4.2 Image of the results of the CQ11 microsatellite assay following electrophoresis on a 2% agarose gel. Culex pipiens f. molestus (250bp) and Cx. pipiens f. pipiens (180bp) controls are clearly shown in Lanes 9 and 10 respectively. Lane 1: HyperLadder IV (BioLine), Lane 2: Culex pipiens f. pipiens, Lane 3: individual with fragments for both Cx. pipiens f. pipiens and Cx. pipiens f. molestus, Lanes 4-8: Culex pipiens f. molestus, Lane 9: Culex pipiens f. molestus control and Lane 10: Culex pipiens f. pipiens control.

Figure 4.3 Maximum parsimony tree of the mtDNA Cytochrome Oxidase I gene (COI, 710bp) of Cx. pipiens s.l. and Cx. torrentium specimens (from Table 3.5). Bootstrap values are indicated above branches. Terminal labels show individual DNA numbers, species name, specimen origin or GenBank accession numbers for published sequences. Specimens labelled in orange were morphologically identified as Cx. pipiens s.l. and further distinguished as f. pipiens (n=10), f. molestus (n=14) and hybrids (n=6) using the CQ11 microsatellite assay of Bahnck & Fonseca (2006). Of these, 16 specimens occur within the torrentium clade together with morphologically identified Cx. torrentium (n=10, marked in Black), Cx. vagans and Cx. sitiens were included as outgroups (marked in blue), while Cx. pipiens f. molestus Greece, Cx. quinquefasciatus_Aberdeen both from colonies and f. molestus_london

vii collected in sewage tunnels in Barking, London were used as positive controls (marked in blue).

Figure 5.1 Schematic diagram of a female mosquito showing the abdomen (a) fully (3/3) blood fed, (b) 2/3 bloodfed, (c) 1/3 bloodfed and (d) non– bloodfed or gravid. Mosquito outline from http://www.pestworldforkids.org/mosquitoes.html

Figure 5.2 A Maximum likelihood tree, with bootstrap values of nodes, of mtDNA Cytochrome B oxidase gene sequences. Terminal labels show DNA Number_species_Common name_degree Bloodfed_location or_Genbank accession numbers. Altogether there are 8 distinct clades: Dog (n=3, Genbank sequences=2), Deer (n=22, Genbank sequences = 6), Goat (n=10, Genbank sequences = 2), Cow (n=2, Genbank sequences= 2), Horse (n=1, Genbank sequences =2), Man (n=1, Genbank sequences=1), Bird (n=6, Genbank sequences =2), Mosquito (n=70, Genbank sequences =1).

Figure 5.2a shows bloodfed mosquito specimens from which no vertebrate hosts could be identified using host-specific primers. All specimens shown here were sequenced using Universal CytB primers and the closest match to the sequences obtained was Amigeres subalbatus (mosquito).

Figure 5.2b shows the hosts identified from bloodfed mosquito specimens. Coloured labels indicate samples sequenced with host specific primer: Orange-Deer (n=1), Green-Goat (n=3), Blue- Bird (n=5), brown- Dog (n=3), Black labels indicate samples sequences with Universal CytB primers (Boakye et al., 1999) when host specific primers did not work.

Figure 6.1 Graph showing the relative proportions of immature stages collected by genera in relation to the amount and type of vegetation found in the larval habitats. (Em = emergent vegetation; Fl = floating vegetation; All = floating and emergent vegetation).

Figure 6.2 Graph showing association of three mosquito genera (Anopheles, Culex and Culiseta) with different larval habitats in five regions of England: Devon (Dev), Norfolk (Nor), Somerset (Som), Suffolk (Suf) and two regions of northern Wales (Anglesey (Ang) and Caernarfonshire (Wales).

Figure 6.3 Relative proportions of mosquito immatures collected in July 2006 by habitat type. Natural habitats (denoted with ‘N’) yielded only 33% of total mosquitoes collected (n=530) and included stream margins, stream pools, ponds, lakes, ground pools and ditches. A total of 1072 specimens were collected in man-made habitats. Artificial habitats included water collections in forklift buckets, tarpaulin sheets, tyres, boats, buckets and troughs.

Figure 6.4 Presence of 5 species of British Anopheles mosquitoes collected in southern England and northern Wales in July and August 2006.

viii Symbols contained in a larger circle indicate several collections within the specified locality. Numbers written in the symbols indicate the total number of specimens collected. Co-ordinates for all collections are available in Chapter 2 Table 2.2. Anopheles algeriensis was collected as larvae in Anglesey; An. atroparvus as resting adults in Norfolk and Kent; An. claviger collected as larvae in Anglesey, Devon and Norfolk and as host seeking adults in Devon and Norfolk; An. daciae collected as larvae in Anglesey, Norfolk and Suffolk and as resting adults in Somerset, Kent and Norfolk; and An. messeae collected as larvae in Devon and Suffolk; as resting adults in Somerset, Kent and Norfolk.

Figure 6.5 Presence of Culex mosquitoes collected in northern Wales and in southern England in this study. Symbols contained in a larger circle indicate several collections within the specified locality. Numbers written in the symbols indicate the total number of specimens collected. Co-ordinates for all collections are available in Chapter 2 Table 2.2. Culex pipiens was collected as larvae in Anglesey, Devon, Norfolk, Somerset, Suffolk and in northern Wales and as resting adults in Anglesey and Somerset; as host seeking adults in Devon. Culex torrentium collected as larvae in Devon, Norfolk, England, in Anglesey and Caernarfonshire in Wales and as resting adults in Somerset.

Figure 6.6 Occurrence map of Culiseta mosquitoes collected in southern England and northern Wales in July and August 2006. Symbols contained in a larger circle indicate several collections within the specified locality. Numbers written in the symbols indicate the total number of specimens collected. Co-ordinates for all collections are available in Chapter 2 Table 2.2. Culiseta annulata collected as larvae in Anglesey; as resting adults in Devon and Somerset and Cs. subochrea collected as larvae in Anglesey; as resting adults in Somerset.

Figure 6.7 Occurrence map of Coquillettidia, Dahliana and Ochlerotatus mosquitoes collected in southern England and northern Wales in July 2006. Symbols contained in a larger circle indicate several collections within the specified locality. Numbers written in the symbols indicate the total number of specimens collected. Co-ordinates for all collections are available in Chapter 2 Table 2.2. Coquillettidia richiardii was collected as resting adults in Somerset and as host seeking adults in Norfolk. Dahliana geniculata was collected as a host- seeking adult in Devon. Ochlerotatus detritus and Oc. leucomelas were collected only as host seeking adults in Anglesey.

List of Tables

Table 1.1 List of British Culicid fauna, showing the 33 reported species and feeding preferences where known (Snow, 1990; Ramsdale & Snow, 1995; Medlock et al., 2005; Linton et al., 2005).

Table 2.1 Preferred larval habitats and adult resting sites for currently recognised British Culicidae (Shute, 1930; Marshall, 1938; Staley, 1940; Nye,

ix 1954; Wallace, 1958; Service, 1968; Rees & Rees, 1989; Snow, 1990; Linton et al., 200a2; Linton et al., 2005; Snow & Medlock, 2008). ? indicates no data are available in the literature.

Table 2.2 The localities, co-ordinates and habitats of mosquitoes collected in this study. R and MT Indicate Resting and Mosquito Magnet Trap ® collections and the rest are larval collections

Table 3.1 Dates and exact collection localities of An. maculipennis s.l. collected across southern England and Wales in June and August 2006. *Indicates immature collections, link-reared to adults; all others were collected as resting adults.

Table 3.2 GenBank accession numbers and locality of 210 ITS2 sequences [An. atroparvus from England and Romania (n=42), An. daciae from England and Romania (n=103) and An. messeae from England, Greece, Romania and Sweden (n=65)] used to design the ITS2 PCR-RFLP assay designed to differentiate the three members of the British Maculipennis Group together with sequences generated in this study.

Table 3.3 Relative proportions of species of the Maculipennis Group (n=711) collected and molecularly identified in five English counties (Devon (n=99), Norfolk (n=402), Somerset (n=93), Suffolk (n=6) and Kent (n=110) and on the Welsh island of Anglesey (n=1) in this study. Anopheles atroparvus (n=111), An. daciae (n=471) and An. messeae (n=129) Total numbers identified by the ITS2 PCR-RFLP assay and by direct sequencing of the ITS2 fragment (numbers sequenced are shown in parenthesis) are shown. *indicates larval collections.

Table 4.1 Morphological characters distinguishing Culex pipiens f. pipiens Linnaeus and Culex pipiens f. molestus Forskål (after Marshall & Staley, 1937).

Table 4.2 Composition of the PCR reagents and thermocycler conditions used in the 20µl reaction for the CQ11 microsatellite assay of Bahnck & Fonseca (2006).

Table 4.3 PCR reaction mix and thermocycler conditions for amplification of the barcoding region of the COI gene (25µl reaction).

Table 4.4 Identification of Cx pipiens s.l. specimens using the assay of Bahnck & Fonseca (2006); Cx. pipiens f. molestus (n=95), Cx. pipiens f. pipiens (n=205), hybrids (n=22).

Table 4.5 A total of 30 specimens (from Table 3.4): Cx. pipiens f. pipiens (n=10), Cx. pipiens f. molestus (n=14) and pipiens x molestus hybrid (n=6) and ten morphologically identified Cx. torrentium were sequenced for COI. Upon analysis, the assay of Shaikevich (2007) distinguished 14 Cx. pipiens f. pipiens individuals, 26 Cx. torrentium individuals and no Cx. pipiens f. molestus or hybrid specimens were detected.

x

Table 5.1 Summary of available data on known host selection and multiple feeds in British mosquitoes (1Service, 1971; 2Snow, 1990; 3Cranston et al., 1987; 4Ramsdale & Snow, 1995; 5Medlock et al., 2005; 6Muirhead- Thompson, 1956; 7Mattingly, 1950; 8Shute, 1933; 9Service, 1969).

Table 5.2 List of habitats, collection dates and co-ordinates (in degree decimal) of sites where all adult blood fed and resting mosquitoes were collected for this study. Resting adults were collected in Devon, Kent, Norfolk and Somerset, while host-seeking adults were collected in Anglesey and Norfolk. *Collections were taken from the low open stables of reindeer and miniature donkeys and ponies in adjacent pens of the Petting Area of Pettits Animal Farm.

Table 5.3 Host-specific primers designed in this study to complement the CytB-F primer (5΄-CCATCCAACATCTCAGCATGATGAA-3΄ of Boakye et al. (1999), showing expected fragment sizes and optimal primer concentrations and annealing temperatures. Bird-F (designed by Y.-M. Linton) was designed to pair with the CytB-R primer of Boakye et al. (1999).

Table 5.4 Optimised PCR master mix and thermocycling conditions for the amplification of vertebrate CytB gene, using host-specific primers designed in this study [after Boakye et al. (1999)]. Reagents including 10x NH4 buffer, 50mM MgCl2 and Taq were from BioLine®.

Table 5.5 List of 280 bloodfed (BF) specimens used to analyse host selection and determine natural parasitic infection. Abdomens of bloodfed mosquitoes were scored visually (Figure 4.1a-d) as follows: 1/3 bloodfed, 2/3 bloodfed, 3/3 bloodfed, gravid and non-bloodfed (non- BF). Gravid and non-bloodfed specimens were included as negative controls.

Table 5.6 Host selection of British mosquitoes (n=166) successfully analysed in this study using species diagnostic primers designed herein and CytB DNA sequences.

Table 5.7 Summary of available data as shown on page 109, Red bold text shows the additional hosts detected in this study section 5.4.3.

Table 6.1 Numbers of larval and adults collected per species in each county, by collection type. Immatures (n=222), resting (n=3) and 8 host-seeking individuals were collected using the Mosquito Magnet® trap in Anglesey. Human landing (n=1), larvae/pupae (n=208) resting (n=106) and host seeking female (n=6) collections were made in Devon. Only Resting (n=108) adults were identified in Kent, Larvae/pupae (n=42), resting (n=403) and host-seeking (n=220) adult collections were made in Norfolk. Only Larvae/pupae (n=24) and resting (n=109) collections were made in Somerset. Larvae/pupae (n=12) were collected in Suffolk and in Wales (n=35).

xi

Acknowledgements

My acknowledgement list is a long one and I could not have completed this without every single one of you.

Thank you:

Canterbury Christ Church University (CCCU) for awarding me the studentship that enabled me to carry out this PhD.

The Natural History Museum (NHM), London, for funding my fieldwork in July 2006.

Dr D.J. Ponsonby (CCCU) and Dr Y.-M. Linton (NHM) for their unwavering support and guidance.

Prof. Georges Dussart, the Chair of my Panel, for being encouraging and an inspiration.

Dr Ooi Eng Eong for setting me on this Culicid journey.

Jackie Trigwell, Alastair Dussart and Phil Buckley for the eye-opening undergraduate field course experience (in 2007 and 2008)!

The Mosquito Group at the NHM: Annette Lee, Fredy Ruiz, Lisa Smith and Magdalena Zaroweicki, you were an amazing bunch to work with!

The MSL lab at the NHM: Johannes Bergsten, Raul Bonal, Miranda Elliot, Tomochika Fuijisawa, Benjamin Isambert, Andie Hall, Alex Martin, Sara Pinzon-Navarro, Anna Papadoupoulou, Joan Pons, April Wardana and Ruth Wild, laboratories are never going to be the same again!

The Sequencing Facility at the NHM: Farah Fatih, Claire Griffin and Julia Llewlyn-Hughes.

Lisa-Anne, Jacqueline Yap-Furley, Jaspal Sunita Kaur-Griffin, Bridget Hodson, Sasibai Kimis, Juanita Choo-Koh, Christina Liew, Germaine Chia-O’Shaugnessy, Swetha Ramachandran and Reshani Iranga Satharasinghe for being there.

Last but not least to all my Family: Mom, Dad, Vik, Mary Monaghan, Marianne, David, Jacob, Marjorie, Grant, Kate, David, Devin, Aunty Marcella, Katrina, Brian, Megan, Maisie, Scott, Carla, Laurel, Lindsay, Genevieve, Aunty Carol, Uncle Martin and Jeremy.

I would like to dedicate this thesis to two very special people: my husband Michael T. Monaghan, my pillar and to Ms G Rosalind J whose determination and perseverance will always be an inspiration to me.

1

Abstract

As early as 1901, ecological and epidemiological studies were conducted to understand malaria transmission in the UK. Unfortunately, following the eradication of malaria after WWII, ecological studies on local mosquito species has been intermittent, leading to a significant gap in knowledge of the current habitat preference, distribution and vector capabilities of the 33 recorded species. This lack of current information makes the assessment of possible transmission of enzoonotic diseases such as Chikungunya and West Nile virus in UK difficult. Thus the overall purpose of this thesis was to facilitate the identification of potential vector species through the documentation and characterisation of the ecology of adult and larval stages, and the host selection of British mosquitoes, in southern England and northern Wales.

A total of 13 out the 33 documented species are assessed in this study. Of which members of the Maculipennis and Pipiens Group comprised the bulk of the adult and immature collections respectively. The development of the ITS2 PCR-RFLP assay in this study allowed the identification of the three members of the Maculipennis Group, which revealed the widespread occurrence of the recently documented An. daciae in almost all localities sampled. While previously published assays discriminating the Pipiens Complex, did not yield congruent results questioning the prior identification methods and the validity of the taxonomic status of its members. In addition, host-specific primers designed herein to determine host selection in local mosquitoes revealed an indiscriminate host selection by An. atroparvus, An. daciae, An. messeae and Cx. pipiens thus indicating their potential role as vectors in the UK.

2

Chapter 1

General Introduction

3

1. General Introduction

1.1 History of Mosquito Systematics

For the first time, following Ross’s (1897) publication of the role of mosquitoes in the transmission of malaria, serious efforts were taken to actively describe as many mosquito species as possible and to identify taxonomic groupings. In 1910, F. V. Theobald published his ‘Monograph of the Culicidae’ proposing the first classification of mosquitoes. This and other early taxonomic studies on the European fauna by F.W. Edwards, resulted in an initial classification of the family Culicidae, which comprised three subfamilies (Edwards, 1932): Culicinae (the “true mosquitoes”) and Dixidae and Chaoboridae (both “midges”) (Figure 1.1). Despite much scientific debate on whether the midges should be included as subfamilies of the family Culicidae, the taxonomic framework of Edwards (1932) was upheld until Knight & Stone (1977) formally removed the Dixidae and Chaoboridae from within the family Culicidae. This key change recognised Anophelinae, Culicinae and Toxorhynchitinae as subfamilies within Culicidae (Stone, 1956) and the reorganisation of the subfamily Culicinae into 10 recognised tribes (Belkin, 1962) resulted in the working framework of mosquito classification used today (Munstermann & Conn, 1997) (Figure 1.2).

The biomedical importance of many mosquitoes and the potential application that phylogenetics could provide in terms of answering questions on vector capacity, diversification of mosquitoes and epidemiology (Sallum et al., 2000; Krzywinski et al., 2001a, b) has resulted in a shift from descriptive taxonomy to natural classification (Zavortink, 1990; Reinert et al., 2004, Shepard et al., 2006). Ross (1951, in Harbach & Kitching, 1998) made the first attempt to create ‘intuitive’ phylogenetic trees based on morphology and species distribution. Despite the biomedical importance of mosquitoes, these hypotheses remained largely unchallenged for fifty years (Harbach & Kitching, 1998). According to Zavortink (1990), natural classification allows for species to be studied in greater detail thus accurately identifying genetic relationships at the generic and sub-generic level. Harbach & Kitching (1998) attempted natural classification based on morphological

4

Culicidae

Culicinae Chaoboridae Dixidae

Culicini

Toxorhynchitiini

Uranotaenia Aedes Sabethes

Culex Theobaldia

Anophelini

Figure 1.1 Classification of the Family Culicidae after Edwards (1932)

5

Culicidae

Culicinae

Culicini Ficalbiini Mansoniini Sabethini Uranotaeniini Aedeomyiini Culisetini

Hodgesiini Orthopodomyiini Aedini Toxorhynchitiini Anophelinae

Anopheles Bironella Chagasia

Figure 1.2 Organogram of Family Culicidae after Belkin (1962) and Harbach & Kitching (1998).

6 characters of larvae, pupae and adults. They found relationships within tribes of subfamily Culicinae to be poorly resolved. This could be explained by hybridisation of characters occurring between 2 or more taxa (Belkin, 1962; Zavortink, 1990), thus implying that mosquitoes within tribes are derived from the same ancestor (Harbach & Kitching, 1998).

One explanation for the lack of effort in mosquito systematics could simply be that formal species descriptions and identification keys have taken precedence due to their biomedical importance in disease transmission (Besansky & Fahey, 1997; Krzywinski et al., 2001a, b). However, the relatively recent advent of sophisticated cladistics methods to analyse morphological characters and the application of molecular techniques to mosquito systematics have catalysed the current heightened levels of activity in Culicid systematics (Harbach & Kitching, 1998), leading to several high-level phylogenies, particularly those dealing with the Tribe Aedini (Reinert, 2000; Reinert et al., 2004; Reinert & Harbach, 2005; Reinert et al., 2006; Reinert et al., 2008; see section 1.1.2).

1.1.1 Internal Systematics of the Subfamily Anophelinae

According to the morphological study of Harbach & Kitching (1998), the subfamily Anophelinae comprises three genera: Anopheles, Bironella and Chagasia. Given the minor reported morphological differences between mosquitoes of genera Bironella and Anopheles (Belkin, 1962; Tenorio, 1977), Sallum et al. (2000) investigated the relationship of the genus Bironella within the subfamily Anophelinae. Based on morphology, Sallum et al. (2000) found that species in genera Anopheles and Bironella comprised three major lineages: Lineage 1 (proposed as the oldest) - Neotropical subgenera Nyssorhynchus and Kerteszia; Lineage 2 – the monophyletic subgenus Cellia (most closely related to lineage 3); and Lineage 3 which was paraphyletic, consisting of the subgenera Anopheles, Lophopodomyia, Stethomyia and the purportedly separate genus Bironella. That Bironella occurred as a subgenus within genus Anopheles (Sallum et al., 2000) contradicted the earlier findings of Harbach & Kitching (1998).

Morphological studies (Harbach & Kitching, 2005) and molecular studies (Krzywinski et al., 2001a, b) agree with the proposed monophyly of the subgenera Nyssorhynchus, Kerteszia and Cellia. However, their findings disagree with the placement of Bironella as a subgenus within Anopheles by Sallum et al. (2000). Using molecular data, Bironella was 7 placed basal to Anopheles, suggesting that it diverged much earlier and provided evidence for its separate generic status (Krzywinski et al., 2001b). In 2002, Sallum et al. re-examined the phylogeny of Anopheles and Bironella using both morphological and molecular characters. Their results still showed Bironella to be a distinct group within the genus Anopheles. As such the relationship of Bironella with respect to the genus Anopheles remains unresolved (Sallum et al., 2002).

Currently, the genus Anopheles includes 459 formally named and more than 50 provisionally designated extant species divided between seven subgenera: Anopheles, Baimaia, Cellia, Kerteszia, Lophopodomyia, Nyssorhynchus and Stethomyia (Harbach & Howard, 2007; updated on Mosquito Taxonomic Inventory website, 2008).

1.1.2 Internal Systematics of the Subfamily Culicinae

The subfamily Culicinae is the largest subfamily in Culicidae, comprising the tribes Culicini and Aedini (Harbach & Kitching, 1998). The tribe Aedini is the largest in subfamily Culicinae consisting of 1,235 recognised species to date (Reinert et al., 2006). According to Belkin (1962), the tribe Aedini appeared not to be a monophyletic group, but more a heterogeneous mixture of species that were superficially similar, making the construction of a natural classification challenging. One approach was to divide the species of the tribe into smaller groups by creating more genera, thus accurately determining phylogenetic relationships (Zavortink, 1990). Reinert (2000) provided morphological evidence based on characters of the male and female genitalia, 4th larval instar and one pupal character to elevate mosquitoes of the subgenus Aedes (Ochlerotatus), to generic status as Ochlerotatus. Of the thirteen British species in the genus Aedes at that time, only two remained in the genus Aedes, with the other eleven being transferred to the genus Ochlerotatus following the work of Reinert (2000) (Table 1.1). Although the results of this study were widely challenged at the time by fellow entomologists (AMCA, 2002), the elevation of Ochlerotatus has since been further supported by additional morphological studies (Reinert et al., 2004) and by molecular data (Spanakos et al., 2006).

Following the papers of Harbach & Kitching (1998) and Reinert (2000), the authors collaborated to re-evaluate the entire internal classification of tribe Aedini using 172 morphological characters of both the adult and larval stages, representing the recognised 12 8 genera and 56 subgenera (Reinert et al., 2004). In some cases, the resulting phylogenies were shown to vary depending on the subsets of data used, character codings and weightings applied. However, some 55 groups were consistently identified as monophyletic regardless of data and weighting used and these robust subgeneric groupings were elevated to full generic status, creating a further 43 new genera. Shepard et al. (2006) maintained that whilst the characters used by Reinert et al. (2004) were diagnostic for species identification, they lacked sufficient depth to be employed in resolving evolutionary relationships. This reorganisation of the tribe Aedini caused so much controversy, especially amongst US mosquito workers who suggested that the evidence was not strong enough for such a radical shake to the taxonomic stability, that an internet based forum was established for the sole purpose of discussing these changes (Walter Reed Biosystematics Unit, 2008). Major concerns included the lack of inclusion of readily available molecular data to confirm these elevations, and the taxonomic instability that such a large number of changes would cause, thus leading to confusion over taxon names particularly amongst non-scientific mosquito-personnel. The most controversial of these is the name change of the Asian tiger mosquito, from Aedes (Stegomyia) albopictus to the almost unrecognisable Stegomyia albopicta (Reinert & Harbach, 2005).

The revision of the internal systematics of the tribe Aedini is still ongoing. Reinert et al. (2006) have recently reevaluated the classification and phylogeny of Finlaya. Based on the morphology of the egg, fourth instar larvae, pupae and adults and on the preferred habitats of the aquatic stages, another 11 subgenera were elevated to generic status; thus resulting in further name changes. The British species Aedes (Finlaya) geniculatus (Olivier, 1919), changed to Ochlerotatus geniculatus (Reinert et al., 2004), has been changed again to Dahliana geniculata, following the most recent review of Finlaya mosquitoes (Reinert et al., 2006). It seems clear that the taxonomic community must clarify these changes by adopting integrated molecular and morphological character assessments before a final resolution is agreed upon.

Based on the number of taxa in the family Culicidae, Zavortink (1990) proposed that the internal classification would comprise 225 genera; by this proposal, the tribe Aedini should comprise circa 87 genera (Reinert et al., 2004). At the moment, 64 separate genera have been recognised in the tribe Aedini (Mosquito Taxonomic Inventory, 2008). The reassessment of this tribe Aedini is a four-phase project aimed to create a natural

9 classification of the tribe (Reinert et al., 2006), as such we can probably expect further changes to the internal systematics of Aedini in the future.

Unlike Aedini, tribe Culicini was found to be a monophyletic group (Harbach & Kitching, 1998) comprising of 4 currently recognised genera: Culex, Deinocerites, Galindomyia and Lutzia (Belkin, 1962). Navarro & Liria (2000) determined the phylogenetic relationships of eighteen Neotropical Culicini species based on 30 characters of the larval mouthparts; their findings supported the proposed monophyly of the tribe by Harbach & Kitching (1998). However, their analysis indicated that the genus Deinocerites arises well within the genus Culex, contradicting the sister-group relationship proposed by Harbach & Kitching (1998). Further, they suggested the reduction of Deinocerites to a subgenus within Culex (Navarro & Liria, 2000). Given the number of taxa occurring within Culex, a number of morphological characteristics have been found to be polymorphic (Harbach & Kitching, 1998) accounting for the poor resolution encountered between subgenera based on morphological assessments alone (Navarro & Liria, 2000).

The genus Culex has a worldwide distribution and comprises more than 762 species in 26 subgenera (Mosquito Taxonomic Inventory, 2008) with Culex pipiens being the nominotypical species for this genus (Linnaeus, 1758, in Harbach et al., 1985). Following the differentiation of Culex torrentium based on the structure of the male phallosome (Martini, 1925), the internal systematics of the genus Culex began to take shape. Sirinivanakarn (1976) initially proposed the subgenus Culex to be split into two groups: Pipiens Group and Sitiens Group. The addition of two Nearctic species allowed for the Pipiens Group to be further divided into 4 subgroups (Figure 1.3). Due to the morphological and genetic overlap between the species within the two subgroups, Dahl (1978) proposed that the Pipiens and Trifilatus subgroups be collapsed. However, based on distinct differences on the male genitalia between the Pipiens and Trifilatus subgroups, this proposition was overturned by Harbach (1988) and the division of subgenus Culex was reverted to the four groups previously listed.

10

Subgenus Culex

Pipiens Sitiens Duttoni Atriceps Group Group Group Group

Theileri Univittatus Trifliatus Pipiens subgroup subgroup subgroup subgroup • pipiens and f. molestus • torrentium • quinquefasciatus

• vagans • pallens • restuans • australicus • globixtus

Figure 1.3 Organogram of the subgenus Culex, following the most recent classification of Harbach (1988).

11

1.2 British Mosquitoes

1.2.1 Species List and Feeding Preferences

Mosquitoes of the British Isles comprise thirty-four recognised species in eight genera: Aedes (3 sp.), Anopheles (6 sp.), Culex (4 sp. plus one form), Dahliana (1 sp.), Culiseta (7 sp.), Coquillettidia (1 sp.), Ochlerotatus (11 sp.) and Orthopodomyia (1 sp.) (Snow, 1990; Ramsdale & Snow, 2000, Linton et al., 2002a; Linton et al., 2005; Reinert et al., 2006) (Table 1.1). Following the eradication of human malaria after WWII (Dobson, 1989) and the subsequent lack of importance of British mosquitoes as vectors, little is known about the current distribution, ecology and feeding behaviour of these mosquitoes. Data gathered from previously published reports on the preferred hosts of British mosquitoes (Table 1.1) show, 28 species feeding on human hosts, whilst only 13 are thought (but not confirmed) to be ornithophilic (Table 1.1). Culiseta annulata is opportunistic in its feeding behaviour, possibly feeding on mammals, man and birds. Culex europaeus is unusual in that it is the only British species to feed primarily on amphibians and reptiles. The host preference of An. daciae and Cs. fumipennis has not yet been determined. Determination of host preferences can provide valuable information on the potential vectorial capacity of these mosquitoes, thus enabling the identification and incrimination potential vectors of animal and human diseases as well as bridge vectors of zoonotic diseases, in the UK.

12

SPECIES HOST PREFERENCE

Aedes cinereus (Meigen, 1818) cattle, mammals, man, birds Aedes geminus Peus 1970 ? Aedes vexans (Meigen, 1830) man

Anopheles algeriensis Theobald, 1903 man Anopheles atroparvus Van Thiel, 1927 man Anopheles claviger (Meigen, 1804) mammals, man Anopheles daciae Linton, Nicolescu & Harbach, 2004 ? Anopheles messeae Falleroni, 1926 man Anopheles plumbeus Stephens, 1828 mammals, man, birds

Coquillettidia richiardii (Ficalbi, 1889) birds, cattle, man

Culex europaeus amphibians, reptiles, man Da Cunha Ramos, Ribeiro & Harrison (2003) Culex modestus Ficalbi, 1889 man Culex pipiens Linnaeus, 1758 (f. pipiens) birds f. molestus Forskål 1775 man Culex torrentium Martini, 1925 birds

Culiseta annulata (Schrank, 1776) man, mammals, birds Culiseta alaskaensis (Ludlow, 1906) man Culiseta fumipennis (Stephens, 1825) ? Culiseta litorea (Shute, 1928) mainly birds, man Culiseta longiareolata (Macquart, 1838) birds Culiseta morsitans (Theobald, 1901) mainly birds, man Culiseta subochrea (Edwards, 1921) man

Dahliana geniculata (Olivier, 1919) man

Ochlerotatus annulipes (Meigen, 1830) mammals, man Ochlerotatus cantans (Meigen, 1830) mainly cattle, also man/birds Ochlerotatus caspius (Pallas, 1771) man Ochlerotatus communis (De Geer, 1776) man Ochlerotatus detritus (Haliday, 1833) mainly cattle, also man/birds Ochlerotatus dorsalis (Meigen, 1830) cattle, man Ochlerotatus flavescens (Muller, 1764) mammals, man Ochlerotatus leucomelas (Meigen, 1804) man Ochlerotatus punctor (Kirby, 1837) mainly cattle, also man/birds Ochlerotatus rusticus (Rossi, 1790) man Ochlerotatus sticticus (Meigen, 1838) man

Orthopodomyia pulcripalpis (Rondani, 1872) birds

Table 1.1 List of British Culicid fauna, showing the 33 reported species and feeding preferences where known (Snow, 1990; Ramsdale & Snow, 1995; Medlock et al., 2005; Linton et al., 2005).

13

1.2.2 Distribution of British Mosquitoes

As part of an investigation into vectors of malaria in Britain, Nuttall et al. (1901) plotted the known distributions of Anopheles mosquitoes in England and Wales and Marshall (1938) provided a comprehensive distribution of the mosquitoes in the U.K. Since then a series of papers have reviewed the distribution of British mosquitoes by genus: Anopheles (Rees & Snow, 1990), Culex (Rees & Snow, 1992), Coquillettidia, Culiseta and Orthopodomyia (Rees & Snow, 1994), Aedes (Rees & Snow, 1995) and Ochlerotatus [Rees & Snow, 1996 (as a subgenus of Aedes)]. The data used to create these maps were obtained from Marshall (1938), Cranston et al. (1987) and the British Mosquito Recording Scheme; however up-to-date distribution data is needed, especially at species level for potential vectors and members of species complexes, for e.g. Maculipennis Complex.

1.2.2.1 Distribution of the subfamily Culicinae in the UK

Prior to publications by Rees & Snow (1990, 1992, 1994, 1995, 1996), no distribution maps for British mosquitoes of the genera Aedes (including Ochlerotatus and Dahliana species), Culex, Orthopodomyia, Coquillettidia and Culiseta had appeared in the literature. Culex species appear to be widespread, with distribution over the coastal areas of south England ranging from Norfolk through to Cornwall and Wales. However, older records for Culex do not differentiate between Cx. pipiens f. pipiens, Cx. pipiens f. molestus or Cx. torrentium and these were all recorded as Cx. pipiens s.l. (Rees & Snow, 1992). The tribe Aedini is one of the largest groups of British mosquitoes comprising 15 species in three subgenera, Aedes, Dahliana and Ochlerotatus (Reinert et al., 2004; Reinert et al., 2006). They range from the south coast of England to as far north as Edinburgh and into the Highlands of Scotland (Rees & Snow, 1992; Rees & Snow, 1994; Rees & Snow, 1995; Rees & Snow, 1996; Snow et al., 1998) (Figure 1.4 a, b, c).

14

1.2.2.2 Distribution of the genus Anopheles in the UK

Six species of Anopheles (subgenus Anopheles) species have been recorded in Britain: An. (An.) algeriensis, An. (An.) claviger and An. (An.) plumbeus and three members of the An. (An.) maculipennis complex: An. atroparvus, An. messeae and An. daciae (Rees & Snow, 1990; Snow, 1990; Snow, 1998; Linton et al., 2002a; Linton et al., 2005) (Figure 1.5 a, b). However, the distribution maps tend to be somewhat biased, reflecting the locality of entomologists as much as that of the species studied (Snow, 1998). As current molecular techniques have improved species level identification, more up-to-date maps reflecting species distribution can and should be generated.

1.2.3 British Mosquito Species Complexes

1.2.3.1 The Anopheles maculipennis Species Complex

Anopheles maculipennis, the historical vector of human malaria in Europe was the first Anopheles species to be exposed as comprising cryptic sibling species (Falleroni, 1926; van Thiel, 1927). Studies to elucidate component members of the group included egg morphology (Falleroni, 1926; Falleroni, 1932; Corradetti, 1934; de Buck & Swellengrebel, 1934a; Hackett & Lewis, 1935; Korvenkontio et al., 1979), hybridisation experiments (de Buck & Swellengrebel, 1934b; Kitzmiller et al., 1967), ecology studies (van Thiel, 1927; de Buck & Swellengrebel, 1934b; Hackett & Missiroli, 1935), larval chaetotaxy (La Face, 1931; Diemer, 1935; Bates, 1939; Buonomini, 1940; Işfan, 1952; Suzzoni-Blatger & Sevin, 1981; Boccolini et al., 1986; Suzzoni-Blatger et al., 1990;

15

Cs. longiareolata Cx. torrentium Cs. litorea a Cx. pipiens.s.l b Cs. fumipennis Cx. europaeus Cs. morsitans Cx. modestus Cs. subochrea

Cs. alaskaensis

Cs. annulata Or. pulcripalpis

Ae. cinereus c Ae. vexans

Da. geniculatus

Oc. annulipes

Oc. cantans

Oc. caspius

Oc. communis Oc. detritus Oc. dorsalis Oc. flavescens

Oc. leucomelas

Oc punctor Oc. rusticus Oc. sticticus

Cq. richiardii

Figure 1.4 Distribution of Culicine mosquitoes in the UK. (a) Distribution of genus Culex (modified from Rees & Snow et al., 1992; Snow et al., 1998); (b) Distribution of genera Culiseta and Orthopodomyia (modified from Rees & Snow, 1994; Snow et al., 1998) and (c) Distribution of the Aedes, Coquillettidia, Dahliana and Ochlerotatus genera (modified from Rees & Snow, 1994; Rees & Snow, 1995; Rees & Snow, 1996; Snow et al., 1998). Distribution data for Ae. geminus was not previously published as detection of this species was done using museum specimens (Medlock & Vaux, 2009).

16

An. algeriensis

An. daciae b An. claviger

a An. atroparvus An. plumbeus

An. messeae

Figure 1.5 Distribution of Anopheles mosquitoes in the UK. (a) Distribution of An. atroparvus, An. daciae and An. messeae (modified from Rees & Snow, 1990; Snow, 1998; Linton et al., 2002a; Linton et al., 2005) and (b) Distribution of An. algeriensis, An. claviger and An. plumbeus (modified from Snow, 1990; Snow, 1998).

17

Deruaz et al., 1991), pupal chaetotaxy (Diemer, 1935; Işfan, 1952), adult morphology (Diemer, 1935; Ungureanu & Shute, 1947; Linton et al., 2003; Sedaghat et al., 2003a; Nicolescu et al., 2004), chromosomes (Frizzi, 1952; Frizzi, 1953; Kitzmiller et al., 1967; Stegnii, 1976; Stegnii & Kabanova, 1976; White, 1978), zymotaxonomy (Korvenkontio et al., 1979; Bullini & Coluzzi, 1982; Cianchi et al., 1987; Jaenson et al., 1986a; Suzzoni-Blatger et al., 1990), cuticular hydrocarbons (Phillips et al., 1990) and, most recently, DNA sequences (Marinucci et al., 1999; Proft et al., 1999; Linton et al., 2001a; Linton et al., 2002a,b,c; Sedaghat et al., 2003a, b; Gordeev et al., 2004; Linton, 2004; Nicolescu et al., 2004; Linton et al., 2005; Gordeev et al., 2005; Linton et al., 2007). These works contributed to the current recognition of eleven Palaearctic members within the Maculipennis Complex: An. artemievi Gordeev et al., An. atroparvus van Thiel, An. beklemishevi Stegnii & Kabanova, An. daciae Nicolescu et al., An. labranchiae Falleroni, An. maculipennis s.s. Meigen, An. martinius Shingarev, An. melanoon Hackett, An. messeae Falleroni, An. persiensis Linton et al. and An. sacharovi Favre (Guy et al., 1976a (review through 1975); White, 1978; de Zulueta et al., 1983; Cianchi et al., 1987; Ribiero et al., 1988; Linton et al., 2002a; Sedaghat et al., 2003b; Nicolescu et al., 2004; Gordeev et al., 2005).

In the UK and Ireland, the Maculipennis Group was thought to consist of two species – An. atroparvus van Thiel and An. messeae Falleroni (Ashe et al., 1991; Ramsdale & Snow, 2000; Linton et al., 2002a). Ecological and biological differences can be used to differentiate the two species found in England. Anopheles messeae can be found breeding preferentially in inland fresh waters that are either stagnant or slow moving, whereas An. atroparvus is more commonly found in brackish-water pools and ditches in coastal regions. Differences in hibernation conditions have also been used to discriminate between the two species. Anopheles atroparvus is normally found hibernating in warm animal shelters where it will periodically feed on the inhabitants, whereas An. messeae is usually found in cold shelters where it undergoes complete hibernation, surviving the winter on its food reserves (Rees & Snow, 1990).

Linton et al. (2002a) showed they were able to discriminate between An. atroparvus and An. messeae in the UK using the second nuclear Internal Transcribed Spacer (ITS2) gene and the mitochondrial Cytochrome c Oxidase I (COI) gene. Based on the sequence variation in the ITS2 and COI gene, Linton et al. (2005) also positively identified An. daciae (a new

18 member of the Maculipennis Group described from Romania), in the Somerset Levels (southwest England) for the first time.

Anopheles daciae is most closely related to and often sympatric with An. messeae (Nicolescu et al., 2004; Linton et al., 2005). Anopheles messeae has a wide distribution being reported from Ireland and Portugal through to China. Although originally described from Romania, An. daciae has since been detected, by correlation of COI data with type specimens of An. daciae (Linton et al., 2005), in England (Linton et al., 2005), Italy, The Netherlands, Former Yugoslavia and Kazakhstan (as An. messeae, Di Luca et al., 2004); suggesting that its distribution is also extensive. There are also conflicting reports of the vector status and biting preferences in An. messeae populations across Europe that may be attributed to An. daciae (Lee et al., 2002; Nicolescu et al., 2004; Linton et al., 2005). Since An. daciae has only so far been found in the Somerset Levels in the UK (Linton et al., 2005), its presence may also be masked by its sibling species, An. messeae, in this country. Currently, little is known about its distribution, ecology and malaria vectorial status in the UK or Europe. The known distribution of the members of the Maculipennis Complex prior to this study is shown in Figure 1.5a.

1.2.3.2 The Culex pipiens species complex

Despite its global distribution and proven roles in disease transmission, the component members of the Cx. pipiens species complex cannot be reliably identified (Smith & Fonseca, 2004). Culex pipiens s.l. can be divided into two main groups: Cx. pipiens, which occurs in temperate regions with a Holarctic distribution and Cx. quinquefasciatus which occurs in subtropical and tropical areas as well as temperate regions (Smith & Fonseca, 2004). In the United States, Cx. quinquefasciatus is a vector of major diseases such as Saint Louis Encephalitis, Japanese Encephalitis and West Nile virus (WNv) Fever.

Culex pipiens has been widely reported in the UK (Figure 1.4a). To date, there are two known forms of this species: Cx. pipiens f. pipiens, which is believed to be ornithophilic and predominantly rural in localised distributions and Cx. pipiens f. molestus, which feeds on humans and is found mainly in semi-urban environments (Smith & Fonseca, 2004). Culex pipiens f. molestus was traditionally differentiated from Cx. pipiens f. pipiens by its autogenous behaviour (ability to lay at least one batch of eggs without taking a blood meal), 19 but recent studies in Portugal showed some populations of genetically confirmed Cx. pipiens f. pipiens also display this character trait (Diaz et al., 2006). A species-diagnostic molecular assay based on microsatellites recently developed to differentiate these forms (Bahnck & Fonseca, 2006) has positively identified the presence of both forms of Cx. pipiens here in the British Isles (A. Curtotti, pers. comm.). Culex pipiens f. pipiens has been incriminated as a vector of WNv in the US and in a recent outbreak of the disease in Romania (as Cx. pipiens s.l.) (Savage et al., 1999). The role of members of the Cx. pipiens complex in disease transmission in Britain remains unknown.

1.3 Mosquitoes as vectors of disease

Although there are 3,508 recognised species of mosquitoes in the world (Mosquito Taxonomic Inventory, 2008), fewer than 100 are vectors of diseases (Tyagi, 2003). Mosquitoes belonging to genera Anopheles, Culex and Stegomyia are of biomedical importance worldwide, transmitting diseases such as Malaria, Dengue Fever, WNv Fever, Yellow Fever and Japanese Encephalitis (Tyagi, 2003).

Transmission of an arthropod-borne (arbo) pathogen or viruses can occur vertically or horizontally. Vertical transmission is the passage of pathogens either directly to offspring within vector populations or between males and females (Mullen & Durden, 2002). However, horizontal transmission is essential for the maintenance of pathogens in the environment and can occur either mechanically or biologically (Mullen & Durden, 2002). Mechanical transmission does not require the pathogen to amplify or undergo any development in the vector. In this case, the role of the arthropod is an extension on contact transmission between vertebrate hosts (Mullen & Durden, 2002), whereas it is essential for the pathogen to undergo development or reproduction in the insect vector (Carn, 1996), when biological transmission occurs (Figure 1.6).

Transmission of mosquito-borne diseases is often horizontal. Mosquitoes are efficient biological vectors of both avian and human malaria, dengue fever and WNv enabling the respective parasites and viruses to reproduce and amplify to high levels prior to transmission (Mullen & Durden, 2002). However, they have also been implicated in the mechanical transmission of the avian poxvirus. By feeding on the lesions and papules of infected birds, the mouthparts of the mosquitoes become infected. As the virus is able to persist for long 20 periods of time on the mosquito mouthparts, they are able to effectively transmit the poxvirus to uninfected birds during feeding (Carn, 1996). Mosquito-borne diseases can be debilitating and lethal diseases to both humans and animals, thus the management of these arbo-diseases lies not only in the eradication of the pathogen, but also with the control of the vector (Wikelski et al., 2004).

1.4 Mosquito-borne Diseases in the UK

1.4.1 Avian Malaria

Avian malaria, caused by intracellular blood parasites belonging to the family Plasmodiidae, comprises of three genera: Plasmodium, Haemoproteus and Leucocytozoon (Atkinson, 1991; Remple, 2004). There are 25 species of Avian Plasmodium. Of these, P. durae, P. elongatum, P. gallinaceum, P. juxtanucleare and P. relictum are most common infections in birds (Atkinson, 1991). Infection by known strains of P. elongatum and P. relictum often are fatal (Remple, 2004). The extent of infection, however, varies depending on host and parasite species (Schrenzel et al., 2003; Remple, 2004; Wood & Cosgrove, 2006). Parasites have been isolated from the following bird groups: Passerines (which includes warblers and sparrows), domestic fowl and raptors (including owls, falcons, hawks & kestrels) (Atkinson, 1991; Schrenzel et al., 2003; Remple, 2004; Tavernier et al., 2005).

Avian malaria is transmitted mainly through the bite of haematophagus arthropods such as Culicoides biting midges, ticks, blackflies and mosquitoes. Mosquito species belonging to the genus Culex have been incriminated as the main vectors of avian Plasmodium parasites (van Riper et al., 1993). Despite their role in human malaria, no natural infections of avian malaria in Anopheles have been reported, although some species have been shown to be susceptible to avian Plasmodium infection (Huff, 1965) and to be good laboratory vectors (Atkinson, 1991).

In the exceptionally hot summer of 1998, avian malaria was reported at zoos across England. All 27 penguins at Marwell Zoo succumbed to the infection and cases were also noted in Bristol Zoo, with 2 recorded mortalities (BBC news online, 1999). Indeed, lack of innate immunity to alien avian malaria in exotic penguin species in British Zoos has resulted

21 in a controlled programme of chemical prophylaxis against malarial parasites (A. Hartley, pers comm).

1.4.2 West Nile virus (WNv)

West Nile virus is a RNA virus within the Japanese Encephalitis serological group of the family Flaviviridae (Buckley et al., 2003; Petersen et al., 2003). It is maintained in the environment through a sylvatic arthropod-bird cycle (Figure 1.7). The virus has been isolated from more than 150 species of wild and domestic birds globally (Van der Meulen et al., 2005). Two lineages of the WNv are known to occur: WNv lineage 1, which circulates amongst the human population and has been isolated from the north eastern United States, Israel, Africa, India and Russia; and WNv lineage 2 from African and Madagascan isolates which is predominantly maintained in the zoonotic cycle (Petersen et al., 2003; Van der Meulen et al., 2005). Genetic analysis of the virus causing an outbreak in North America in 1999, which resulted in deaths of many native and exotic birds, showed it to be similar to the strain found in Israel though it differed from any strains that were isolated prior to the outbreak (Lanicotti et al., 1999).

Bird species of the order Passeriformes (which includes warblers and sparrows) are the most susceptible to the WNv and act as highly competent reservoir hosts, as they are able to develop the highest amount of virus in their blood (viraemia) (Peterson et al., 2004; Van der Meulen et al., 2005). Bird species from the order Chadriiformes as well domestic geese are highly susceptible to infection and disease (Van der Meulen et al., 2005). The extent of infection however, can range from benign, in many European birds, to universally fatal (Peterson et al., 2004). This could be attributed to the age of the birds and genetic variability of the birds and virulence of the viral strains (Van der Meulen et al., 2005).

Mosquitoes are the primary arthropod vectors incriminated in the transmission of the WNv. The virus has been isolated from at least 43 mosquito species; most belonging to the genus Culex (Hubalek & Halouzka, 1999). The WNv has also been isolated from An. maculipennis s.l. in Portugal in 1971 (Filipe, 1972 in Esteves et al., 2005), Culex modestus in southern France (Hannoun et al., 1964) and Culex pipiens s.l. in Romania (Tsai et al., 1998). Currently in Europe, the main vectors of WNv are Cx. modestus, Cx. pipiens and Cq. richardii (Higgs et al., 2004). Aedes cinereus, Ae. vexans, An. maculipennis, Oc. cantans and 22

Oc. caspius have been linked to WNV transmission in other parts of Europe and USA (Tsai et al., 1998; Higgs et al., 2004; Fyodorova et al., 2006). It has long been suspected that migratory birds play a significant role in the transmission of WNv to new regions (Rappole et al., 2000). According to Medlock et al. (2005), outbreaks of WNv in southern Europe have been attributed to the introduction of the virus from infected African migratory birds to the local mosquito population. Higgs et al. (2004) suggested that based on the predicted changes in climate and increased movement of livestock and man that more virulent strains of the WNv could establish in Europe in the future. They also suggested that the range of the virus could extend northwards, with a possibility of introduction into the UK.

1.4.3 Tahyna and Inkoo Viruses

Both the Tahyna and Inkoo viruses belong to the family Bunyaviridae and are known to occur in Western Europe (Hubalek & Halouzka, 1996). Of the two, Tahyna is the more widespread (Gould et al., 2006); reported in France, Russia (L’vov et al., 1972), Italy, Spain, Germany, Austria (Pilaski & Mackenstein, 1985), Sweden (Lundström, 1999), Finland and Norway. It was first isolated in Ae. vexans and Oc. caspius (Bardos & Danielova, 1959); both of which occur in the UK. Although found mainly in non-human hosts such as horses, reindeer and rabbits, human cases of the Tahyna virus have also been reported (Pilaski, 1987; Lundström, 1999). While no reports of human incidence of the virus have been made in the UK, the presence of Tahyna virus was detected (Chastel et al. 1985 cited in Ramsdale and Gunn, 2005) in Devon in two species of rodents. Although this was a single observation in 1985, it suggests the virus was being transmitted among the local mosquitoes; however, virus isolation from British mosquitoes has not been documented.

The Inkoo virus, though less prevalent in Europe, is primarily transmitted by Oc. communis (Hubalek & Halouzka, 1996). It was reported in Finland and subsequently in Sweden (Francy et al., 1989), Norway (Traavik, 1979 in Hubalek & Halouzka, 1996) and Russia; occurring in cows, reindeer as well as humans (Hubalek & Halouzka, 1996; Ramsdale & Gunn, 2005). While circulating primarily in non-human mammals, Ramsdale & Gunn (2005) have suggested the possibility of transmission of both Inkoo and Tahyna to ground- nesting birds in the UK. The viruses could then be maintained in the environment through the local bird populations. Given the presence of all three vector species and the detection of

23

Tahyna in UK, low-level transmission as well as enzoonotic transmission of both viruses is a possibility.

1.4.4 Sindbis Virus

Originally isolated from a Culex mosquito in Cairo, Egypt in 1952 (Nikolassen, 1989), the Sindbis virus (Alphavirus) is now understood to be part of a complex of 5 viruses (Sindbis complex) of which infection with Sindbis and Sindbis-like viruses have been reported in Russia, Sweden, Finland as well as eastern Europe (Nikolassen et al., 1984; Hubalek & Halouzka, 1996; Lundström et al., 2001). The virus has been isolated primarily from Cx. torrentium (serving as a principal vector in Sweden, Lundström et al., 2001), although Cx. pipiens, Cs. morsitans and Ae. cinereus are also thought to be secondary vectors of Sindbis (Nikolassen, 1989; Lundström et al., 1990a, b; Lundström et al., 2001), with birds (mainly Passerines) acting as the main reservoir host for this virus (Lundström, 1999; Lundström et al., 2001).

1.4.5 Arbo-disease transmission in the UK

While these arboviruses are prevalent in Europe, only a suspected low-level transmission of WNv, Sindbis and Tahyna have been suggested in the UK. This could be due to the presence of the European vectors such as Culex pipiens, Oc. communis, Oc. caspius, and Ae. vexans in Britain. As the British Isles form part of the migratory route of many species of birds between Africa and North America, there is a real chance of any of the above- mentioned pathogens being introduced into the UK (Higgs et al., 2004). Buckley et al. (2003) showed seroconversion to the WNv and Sindbis in native British birds and have detected antibodies to the virus in both migratory and native birds; suggesting that at least some species of native British birds have already been exposed to the virus. No local human or animal cases have been detected thus far (DEFRA, 2008), so if the virus is indeed currently present in the UK, it is only cycling within birds. Based on both vector status in the US and Europe and known feeding behaviour, a total of 11 species and 2 species complexes have been identified as potential vectors of the WNv in the UK (Higgs et al., 2004; Medlock et al., 2005): Aedes cinereus, Ae. vexans, An. plumbeus, Cs. annulata, Cs. morsitans, Cs. litorea, Cx. modestus, Cq. richardii, Oc. cantans, Oc. caspius, Oc. detritus, Oc. punctor, An. maculipennis s.l. and Cx. pipiens s.l. Of these, Ae. cinereus and Cx. pipiens can also transmit 24 both Sindbis and Tahyna. Susceptibility of British birds to infection by other arboviruses and their ability to maintain and replicate them and the competency of British mosquito vectors to transmit them to other vertebrate hosts is still uncertain (Gould et al., 2006).

Aside from the presence of vectors and reservoir hosts, the transmission of disease is also dependent on changes in climatic conditions and the herd immunity of a population. Global temperatures are expected to increase by at least 6°C by the end of the 21st century (Meteorological Office UK, 2008). This concomitant increase in humidity and alteration of rainfall patterns are predicted to be conducive to the spread of both vectors and pathogens outside their natural ranges (Khasnis et al., 2005; Haines et al., 2006). For example, WNv or Sindbis virus could become fully established in the UK, given the suspected low-level transmission of these viruses within the bird population of the UK (Buckley et al., 2003). Potential bridge vectors present in the area could then spread the viruses into the local human population where immunity to these viruses is low. To circumvent this transmission cycle, a greater knowledge of mosquito species present as well as their selection on hosts is essential.

In 2004, England’s Chief Medical Officer (CMO) outlined a prevention and control plan for the possible introduction of West Nile virus into the UK (Department of Health, 2004). Although the threat of arthropod-borne diseases, such as WNv and Chikungunya, entering the UK is low, the CMO reiterated the need for constant vigilance and a strong surveillance system for patients, birds and vectors. Unfortunately, the current knowledge on the ecology and distribution of British mosquitoes does not allow for the establishment of proper surveillance systems. It is hoped that data generated in this study could help to fill in some knowledge gaps, providing comprehensive baseline information on the current distribution, ecology and host selection of mosquitoes in southern England and Wales.

1.5 Overall project aims

This study addresses the habitat preference of British mosquitoes sampled as well as determining the presence of members of species complexes and elucidating their role in disease transmission by:

a) Ascertaining species of mosquitoes present and their habitat preferences (larval habitats, adult resting) in southern England and Wales, 25

b) Ascertaining the presence of members of the Anopheles maculipennis and Culex pipiens species complexes in southern England and developing species-specific molecular diagnostic tools to facilitate accurate identification, and c) Identifying host selection of British mosquitoes by developing a CytB PCR assay and thereby identifying potential vectors and bridge vectors of arbo-diseases

26

Chapter 2

Field collection of British Mosquitoes

27

2 Field Collection of British Mosquitoes

2.1 Introduction

2.1.1 British mosquitoes

In 1918 and 1920, Lang published articles detailing aspects of both the morphology and distribution of the 20 known mosquito species recorded in England at that time. These included: Anopheles maculipennis Meigen, An. claviger (Meigen) (as An. bifurcatus), An. plumbeus Stephens, Aedes cinereus (Meigen), Ae. vexans (Meigen) (as Ochlerotatus vexans), Oc. caspius (Pallas), Oc. dorsalis (Meigen) (as Oc. curriei), Oc. cantans (Meigen) (as Oc. waterhousei), Oc. annulipes (Meigen), Oc. detritus (Haliday), Oc. punctor (Kirby) (as Oc. nemorosus), Oc. rusticus (Rossi), Dahliana geniculata (Olivier) (as Finlaya geniculata), Coquillettidia richiardii (Ficalbi) (as Taeniorhynchus richiardii), Culiseta annulata (Shrank) (as Theobaldia annulata), Cs. morsitans (Theobald) (as Culicella morsitans), Cs. fumipennis (Stephens) (as Culicella fumipennis), Culex pipiens Linneaus, Cx. territans Walker (as Cx. apicalis) and Orthopodomyia pulcripalpis (Rondani) (as Orthopodomyia albionensis). Close scrutiny of Anopheles maculipennis across Europe at the time, revealed differences in egg morphology (Falleroni, 1926; van Thiel, 1927) and two species, An. atroparvus and An. messeae were first reported in the UK in 1934 (in Marshall, 1938). Marshall (1938) also updated these earlier works and produced comprehensive descriptions of morphology, ecology and distributions of British mosquitoes, including data on larval habitats and methods of overwintering (diapause) (Table 5.1). This publication added a further nine new species records for UK as follows: An. algeriensis Theobald, Oc. sticticus (Meigen) (as Aedes sticticus), Oc. communis (De Geer) (as Aedes communis), Oc. leucomelas (Meigen) (as Aedes leucomelas), Oc. flavescens (Müller) (as Aedes flavescens), Cs. subochrea (Edwards) (as Theobaldia subochrea), Cs. alaskaensis (Ludlow) (as Theobaldia alaskaensis), Cs. litorea (Shute) (as Theobaldia litorea) and Cx. molestus Forskål, increasing the British mosquitoes to 30.

28

Species Larval Habitat Adult behaviour Environment

Aedes cinereus Flooded margins of permanent ponds Outdoor resting Rural Aedes geminus Freshwater ? ? Aedes vexans Open, unshaded pools of temporary water, flooded grasslands Rural (fresh/brackish)

Anopheles algeriensis Shallow freshwater pools and swamps in calcarious fenland Outdoor resting Rural Anopheles atroparvus Brackish water pools and ditches in salt marshes Indoor resting Rural/ Domestic Anopheles claviger Along shady margins of ponds & lakes; fresh/brackish water Outdoor/animal shelters Rural Anopheles daciae ? Animal shelters Rural Anopheles messeae Fresh water stream margins and ditches Animal shelters/ Indoor resting Rural/ Domestic Anopheles plumbeus Treeholes (rots and pans); fresh water ? Arboreal

Coquillettidia richiardii Permanent ponds, larvae attached to roots and stems of Indoors & outdoors; adults fly late in Rural/Domestic aquatic plants, including Typha, Acorus, Glyceria & the evening Ranunculus spp.

Culex europaeus Cool freshwater, also brackish water ? ? Culex modestus Fresh / brackish water ? Rural Culex pipiens (f. Subterranean freshwater Indoor/ cellars Rural/Domestic molestus) Culex pipiens (f. Shallow ground pools, artificial containers with rainwater Farm shelters/Indoors Rural/domestic pipiens) Culex torrentium Predominantly artificial containers, occasionally tree holes; Farmland Rural/Arboreal fresh water

Culiseta alaskaensis Freshwater ? ? Culiseta annulata Shallow flooded grassland & woodland localities, ditches, Rests indoors (cellars); continues to Rural/Domestic sometimes subterranean; artificial containers; fresh/brackish feed all through the year water Culiseta fumipennis Temporary woodland pools, or edges of weedy permanent ? Rural open pools Culiseta litorea Coastal species, open sunlit slightly brackish pools ? Rural 29

Species Larval Habitat Adult behaviour Environment

Culiseta longiareolata Foul brackish water pools, fresh water butts ? ? Culiseta morsitans Small permanent woodland pools; one record in treeholes Outdoor resting on vegetation; Rural overwinter as larvae Culiseta subochrea Shallow flooded grassland & woodland localities & ditches; Indoor resting; autogenous Rural/Domestic artificial containers (fresh/brackish)

Dahliana geniculata Treeholes (rots and pans) Outdoor resting Arboreal Ochlerotatus annulipes Open / partially shaded ditches and pools and depressions in ? Rural marshy land Ochlerotatus cantans Densely shaded temporary woodland pools, roadside ditches Outdoor resting in low vegetation Rural Ochlerotatus caspius Brackish water pools, salt marshes Outdoor resting Rural Ochlerotatus communis Temporary woodland pools Indoor resting Domestic Ochlerotatus detritus Brackish water pools, salt marshes Indoor resting Rural

Ochlerotatus dorsalis Temporary pools of fresh/Brackish water pools ? Rural Ochlerotatus flavescens Temporary pools of fresh/Brackish water ? Rural Ochlerotatus Temporary and permanent pools of fresh/brackish water ? Rural leucomelas Ochlerotatus punctor Ditches and temporary woodland depressions, mainly in Indoor resting Rural/Domestic areas with acidic, sandy/gravely soils Ochlerotatus rusticus Ditches and shaded temporary pools in deciduous woodlands Outdoor resting Rural with plenty of leaf litter Ochlerotatus sticticus Temporary woodland pools of open water ? Rural

Orthopodomyia Tree holes (rots and pans) Resting in rot holes or pans of old Arboreal pulcripalpis beech trees

Table 2.1 Preferred larval habitats and adult resting sites for currently recognised British Culicidae (Shute, 1930; Marshall, 1938; Staley, 1940; Nye, 1954; Wallace, 1958; Service, 1968; Rees & Rees, 1989; Snow, 1990; Linton et al., 200a2; Linton et al., 2005; Snow & Medlock, 2008). ? Indicates no data are available in the literature.

30

The first record of Cs. longiareolata in southern England (Staley, 1940), brought the total number of UK mosquito species to 31. In summer 1945, Culex modestus Ficalbi was reported from three separate larval collections in Portsmouth, Gosport and on Hayling Island (all within a 5km radius) (Marshall, 1945). Field specimens of Culex torrentium Martini was first reported in the UK by Mattingly (1951), although by revisiting museum specimens held in the Natural History Museum, London (BMNH), Service confirmed the presence of the species some 50 years prior to this record (in Gillies & Gubbins, 1982). Thus, by 1951, 33 species were recorded in the British Isles.

The proposal of Cx. molestus as a separate species from Cx. pipiens sparked much debate amongst UK entomologists (Marshall & Staley, 1937; Barr, 1957; Stone et al., 1959). It was initially differentiated from the closely related Cx. pipiens, based on its adult morphology and biting behaviour (Forskål, 1775, in Harbach et al., 1984) and subsequently differences in larval siphonal index and in aspects of both male and female morphology were detailed by Marshall & Staley (1937) (see Chapter 4, Table 4.1). Following the neotype designations of Cx. molestus (Harbach et al., 1984) and Cx. pipiens (Harbach et al., 1985), Harbach et al. (1984) proposed that due to the lack of distinctive morphological characters between the two species, Cx. molestus was to be considered as a physiological variant of Cx. pipiens. Thus together with the two forms of Cx. pipiens (Cx. pipiens f. pipiens and Cx. pipiens f. molestus) 32 species were included in the morphological identification keys of British mosquitoes by Cranston et al. (1987) and Snow (1990).

No additional taxa or taxonomic changes were noted in the review paper of Snow et al. (1998) and the British species list appeared to be stabilising. However, the re-elevation of the subgenus Ochlerotatus to generic status (Reinert et al., 2000), resulted in generic nomenclatural changes to all former British Aedes, except Ae. cinereus and Ae. vexans. By careful examination of specimens of Cx. territans Walker from Portugal against those from the type locality in USA, Da Cunha Ramos et al. (2003) determined that the European taxa were in fact a distinct species: Culex (Neoculex) europaeus Da Cunha Ramos, Ribiero & Harrison, thus it is now accepted that all previous records of Cx. territans in Europe actually refer to Cx. europaeus. Following further revisions on the Tribe Aedini, Ae. geniculatus (changed to Oc. geniculatus, after Reinert et al., 2004), was reclassified as Dahliana geniculata (Reinert et al., 2006).

31

With the recent description of the newest member to the Maculipennis Group, An. daciae Linton, Nicolescu & Harbach from Romania (Nicolescu et al., 2004), Linton et al. (2005) retrospectively detected this particular species in collections made in the Somerset Levels in 2001. Similarly the reappraisal of male genitalia of Ae. cinereus museum specimens by Medlock & Vaux (2009) revealed the presence of morpholigcal twin species Ae. geminus in the UK. It is suspected that the fairly recent description of Ae. geminus (Peus, 1970 cited in Medlock & Vaux, 2009) coupled with its morphological similarity to Ae.cinereus could have masked the presence Ae. geminus in the UK (Medlock & Vaux, 2009) which is now thought to have been in Britain as early as the 1920s (Medlock & Vaux, 2009). Recently identified specimens from Devon have confirmed its continued presence in Britain (Medlock, J. pers comm.). Thus with the addition of these two species, the British mosquito taxa currently stands at 34 species (Table 2.1).

2.1.2 Ecology of British mosquitoes

The ecological niches of the 34 British mosquitoes exhibit marked differences in both immature and adult ecology. Aquatic niches are particularly diverse. For example, three species of endemic mosquitoes, Anopheles plumbeus, Dahliana geniculata and Orthopodomyia pulcripalpis, rear out in tree holes (Snow, 1990; Snow & Medlock, 2006), while aquatic stages of An. atroparvus and Ochlerotatus species such as Oc. caspius and Oc. detritus, thrive in brackish waters (Cranston et al., 1987; Medlock & Snow, 2006). Immature stages of Cx. pipiens and Cx. torrentium are often found in artificial containers, while those of An. claviger, Cq. richiardii, Cs. fumipennis, Cs. litorea, Cs. morsitans and Cx. europaeus can be found in groundpools, ponds and streams often amongst dense vegetation (Medlock & Snow, 2006; Snow & Medlock, 2008). Larvae of Cq. richiardii have been known survive for long periods of time below the surface of the water (Cranston et al., 1987), obtaining oxygen by piercing the roots of submerged plants such as reed mace (Typhya sp.) and sweet mace (Acora sp.) (Snow & Medlock, 2008). Adult mosquitoes can either be found resting indoors in houses or in animal shelters as observed for An. atroparvus, An. messeae; Cs. annulata, Cx. pipiens and Cx. torrentium (Cranston et al., 1987; Snow, 1990), while adults of Cs. annulata, Cx. pipiens, Da. geniculata, Oc. cantans, Oc. caspius and Oc. detritus are usually found resting outdoors amongst vegetation (Snow, 1990).

32

Given the temperate climate of the British Isles, all local species of mosquitoes overwinter. Eggs of Ochlerotatus are generally resistant to desiccation (Marshall, 1938), therefore species from this genus normally overwinter as eggs (Snow, 1990). This holds true for species such as Oc. annulipes, Oc. cantans, Oc. caspius, Oc. detritus, Oc. punctor, Oc. rusticus, Oc. sticticus, including Cs. morsitans (Cranston et al., 1987). Other British taxa, including An. algeriensis, An. claviger, An. plumbeus, Cq. richiardii, Da. geniculata and Or. pulcripalpis overwinter as larvae (Cranston et al., 1987; Snow, 1990; Snow & Medlock, 2008). Anopheles atroparvus and An. messeae overwinter as fertilised females (Snow, 1990), although An. atroparvus is known to periodically break out of hibernation to feed on the occupants of the shelter. Reports have shown the susceptibility of An. atroparvus and An. messeae to a short day photoperiod which induces ovarian diapause (Cranston et al., 1987), similar to that observed in females of Cx. pipiens (Clements, 1963 in Cranston et al., 1987). Hibernation is not noted in Cs. annulata, where no effect of photoperiod on diapause was also observed (Service, 1968 in Cranston et al., 1987).

2.2 Materials and Methods

This chapter describes in detail materials and methods used in the collection of larval and pupal stages as well as resting and host seeking adult mosquitoes.

The field collections were carried out according to the standardised Natural History Museum procedure, following the recommendations of Belkin et al. (1962). Field data recorded for each capture included: collection type (e.g. immature, resting, human landing etc), assessment of vegetation present in larval habitat (floating/emergent and abundance of algae) and the potential hosts present (Figure 2.1). Data from each collection site were later correlated with the mosquito identifications (Figure 2.2) in order to more fully characterize the ecological parameters of mosquito species collected.

33

Collection No. Nearest Town Date

Specific Locality Time Country Province Latitude/Longitude Collector(s)

Second Administrative Division Elevation Organisation

COLLECTION TYPE ENVIRONMENT LARVAL HABITAT WATER: Immature Woodland Pond – Lake Permanent Resting - Evergreen Forest Ground Pool Temporary House Deciduous Forest Swamp Animal Shelter Grassland Marshy Depression WATER MOVEMENT Cave Island Stream Margin Stagnant Tree Hole Swamp Stream Pool Slow Vegetation Salt Marsh Rock Pool Moderate Other: ______Beach Seepage - Spring Fast Biting/ Landing - Orchard - Plantation Ditch Human Cultivated Field: ______Well SALINITY Animal: ______Urban Artificial Container Fresh Net Village Hoof Print Brackish Light Trap: ______Other: ______Rut Bait Trap Other: ______TURBIDITY Swarming ENVIRONMENTAL MODIFIERS Clear At Light Primary ALGAE Coloured Other: ______Secondary Filamentous Turbid Green Polluted TERRAIN Agriculture Blue-Green Mountain Pasture Brown PHYSICAL FACTORS Hill Grove/Plantation: ______Other: ______pH Valley Other: ______Conductivity Plateau ALGAL DENSITY Temperature (ºC) Plain WIND None TDS None Scarce DISTANCE FROM HOMES Light Moderate AQUATIC VEGETATION m Gusts Abundant Submerged Strong Floating SKY DIMENSIONS OF SITE Emergent Clear HEIGHT ABOVE GROUND m X m Submerged and Floating Partly Cloudy m Submerged and Emergent Overcast Depth m Floating and Emergent Fog SHADE All Types Mist None QUANTITY OF AQUATIC VEGETATION Light Rain Partial None Heavy Rain Heavy Scarce Moderate HOST Abundant Human Horse Cow REMARKS

Figure 2.1: Collection form.

34

Collection Number EN12 Country England Number Le Pe Sex Identification / Notes -1 √ ♂ Anopheles (An.) claviger (Pinned) -2 √ √ ♀ Culex spp. (in EtOH)

Figure 2.2: The reverse side of the field collection forms (see Figure 5.1) showing an example of the detailed notes kept of each individual larval rearing.

35

County Nearest Town Co-ordinates (Degree decimal) Habitat Anglesey Benllech 53.2011838'N, 04.3714482'W Ditch 53.3153544'N, 04.2379822'W Stream pool 53.3173278'N, 04.2382357'W Nr horse stablesMT Bath Tub Bucket Water trough 53.3178668'N, 04.2382639'W Bath Tub Water trough Brynteg 53.3065719'N, 04.2649949'W Bath Tub Llangferfechan 53.2489805'N, 03.9958504'W Bucket Water trough Newborough 53.1635733'N, 04.3316682'W Bucket Rim 53.1638428'N, 04.3316828'W ShedR, MT Pond, Bird WorldR, MT Penraeth 53.2942853'N, 04.1911069'W Bucket Plastic Barrel Mallraeth 53.2969830'N, 04.1910949'W Tarpaulin Sheet 53.2000146'N, 04.3396364'W Tyre 53.2009233'N, 04.3709845'W Ditch 53.2017891'N, 04.3454238'W Stream pool 53.2485910'N, 03.9976312'W Ditch 53.2728030'N, 04.2438583'W Ditch 53.2736646'N, 04.2458535'W Ground Pool 53.2738329'N, 04.2464624'W Ditch Caernarfonshire Betws-y-Coed 53.0540031'N, 03.8307297’W Water trough 53.0552657'N, 03.8304847’W BBQ Bowl 53.0555421'N, 03.8300487’W Bucket 53.0555731'N, 03.8294598’W Bucket Devon Belstone 50.7274548'N, 03.9508172'W Ditch 50.7358999'N, 03.9674681'W Horse StablesR Bucket Plastic Barrel 36

County Nearest Town Co-ordinates (Degree decimal) Habitat Devon Bridestowe 50.6830244'N, 04.0956511'W Bucket 50.6837282'N, 04.0965319'W Human 50.6839079'N, 04.0965399'W Bucket 50.6839978'N, 04.0965439'W Black Bins 50.6840673'N, 04.0976797'W Bucket Forklift bucket 50.6842216'N, 04.0991025'W Bucket Cheriton Bishop CP 50.7114499'N, 03.7098920'W Boat 50.7130395'N, 03.7422498'W Ground Pool Exminster Marshes 50.6742739'N, 03.4679099'W Stream Margin 50.6752617'N, 03.4751598'W Brick ShelterR 50.6756321'N, 03.4743221'W Brick ShelterR 50.6765270'N, 03.4675556'W Stream Margin 50.6773582'N, 03.4729606'W Stream Margin 50.6784496'N, 03.4720039'W Ditch 50.6786812'N, 03.4679059'W Ditch 50.6789962'N, 03.4714547'W Ditch Launceston 50.5812099'N, 04.3317238'W Pond Moretonhampstead 50.6463599'N, 03.8149006'W Boat 50.6466229'N, 03.8153351'W Water trough Near Chicken coopMT 50.6488266'N, 03.8124491'W Pond 50.6469605'N, 03.8167628'W Horse StablesMT Okehampton 50.6831143'N, 04.0956551'W Stream pool 50.7291615'N, 04.0607206'W Lake 50.7335964'N, 03.9971353'W Stream pool St Erney 50.4011439'N, 04.2806501'W Tyre 50.4026551'N, 04.2815672'W Pond Trevollard 50.3859807'N, 04.2555818'W Ground Pool Two Bridges (Dartmoor Forest) 50.5588627'N, 03.9685579'W Bath Tub Kent Cliffe Marshes 51.4537556'N, 00.4782834’E Brick shelter 1R 51.4671890'N, 00.4895240’E Salt Lane BunkerR 37

County Nearest Town Co-ordinates (Degree decimal) Habitat Kent 51.4783052'N, 00.4827834’E Sheep corralR Norfolk Ranworth 52.6814994'N, 01.4881677’E Boat Reedham 52.4656741'N, 01.5704780'E Petting Area 1MT 52.5659393'N, 01.5805442'E Pond 52.5659609'N, 01.5768554'E Goat StablesR Petting areaR, MT Reindeer StablesR Rhea StablesR 52.5661226'N, 01.5774591'E Petting area 2MT Stream pool 52.5661404'N, 01.5768701'E Horse/donkey stableR Petting areaR West Somerton 52.7214148'N, 01.6559478'E Stream Margin 52.7215319'N, 01.6550687'E Boat Stream Margin Somerset Godney 51.1816487'N, 02.7273050'W Pillbox 1R Ground Pool 51.1817652'N, 02.7230142'W Horse StablesR 51.1822719'N, 02.7283164'W Pillbox 2R Highbridge 51.1873104'N, 03.0173270’W Ditch Otterhampton 51.2106327'N, 03.0344497’W Lake Stockland Bristol 51.1879387'N, 03.0777330’W Ditch Suffolk Beccles 52.4658536'N, 01.5704926’E Stream pool 52.4677069'N, 01.5716738’E Ditch 52.4693843'N, 01.5697476’E Ditch Oulton 52.4682112'N, 01.6927924’E Ditch 52.4683907'N, 01.6928074’E Ditch

Table 2.2 The localities, co-ordinates and habitats of mosquitoes collected in this study. R and MT Indicate Resting and Mosquito Magnet Trap ® collections and the rest are larval collections.

38

Figure 2.3: Collection of aquatic stages in natural habitats (a, b & d) streams and (e) a pond and artificial habitats including (c) disused bathtub, (f) moored boat and (g) fire buckets. Methods of collection for aquatic stages are shown in 2.3d.

39

2.2.1 Collection of immature stages

Larvae and pupae were collected from breeding sites - both natural and artificial (Figure 2.3) - (Table 2.2) using standardised mosquito dippers (Figure 2.3d, BioQuip, LA, USA). Larvae and pupae were carefully isolated using wide-mouthed plastic pipettes and transferred to labelled plastic Whirl-Pak® (Nasco) bags, containing ample water and fine debris from the original breeding site. Bags were sealed for transportation. Once in the laboratory, the contents of each Whirl-pak® were individually transferred into bowls labelled with the corresponding collection number (e.g. EN101) and the immatures were fed Tetramin® powdered baby fish food once daily. Immatures were reared collectively until reaching the fourth instar, whereby each specimen was individually reared.

2.2.1.1 Individual rearing

Individual rearing were carried out from the fourth larval instar and on those collected as pupae in the field. All fourth instar larvae / pupae were transferred individually into plastic vials containing 2-3 cm of original habitat water. Plastic vials were marked with the collection number using a wax pencil. Vials containing isolated larvae, marked with collection numbers, were examined twice a day - early morning and late afternoon. Larval skins were removed with an applicator stick and transferred to a small glass storage vial with 80% alcohol and attached to the rearing vial containing the pupa using an elastic band. A lid was placed on the rearing vial to prevent escape of the adult mosquito after emergence.

The pupal exuviae was then placed into the same vial as the associated larval skin and both the adult and linked exuviae were labelled with the same unique number (e.g. EN101-07, where EN101 indicates English collection 101 and -07 the individual in that collection). Viable adults were pinned for taxonomic studies as described below and associated numbers added to the pinned specimen so both exuviae and specimen were linked. Pupae, larvae and partially emerged adults that died during rearing were preserved in 80% ethanol for DNA studies.

40

2.2.1.2 Preparation of link-reared adults

Adults were transferred from the rearing vials to killing tubes containing plaster infused with ethyl acetate. The adult mosquito and its corresponding identification label were transferred from the killing tube onto clean white card. Using fine forceps, the adult was picked up by one leg and placed on an elevated surface (e.g. postal box) with a white background. The specimen was oriented with proboscis facing right and moved to the edge of the elevated surface with its legs projecting beyond the edge. A card point was fixed an appropriate distance from the head of an insect pin and a tiny droplet of Ambroid® cement put on the upper apical angle of the point. Holding the pin so that the point is upside down, the droplet of glue was gently touched to the thorax of the mosquito. Final orientation of the mosquito on the upper surface of the point was with the left side up, head facing left and the legs extended toward the pin. This orientation protects the specimen from damage and corresponds to the preferred orientation of illustrations in key taxonomic publications (e.g. Harbach & Knight, 1980). The label with the collection and rearing number was attached to the pin. Once pinned, specimens were kept in postal boxes in zip-sealed large plastic bags, to avoid destruction by insectivorous pests. Pinned insect also underwent freezing at -80oC for 5 days after processing to kill fungal spores and bacteria, which can also damage the specimens. Rearing records (Figure 2.2) are maintained on the back of the original field collection records.

2.2.2 Field collections of adult mosquitoes

Collections of resting mosquitoes were carried out manually (Table 2.2) inside animal stables, in abandoned war bunkers and from the walls and ceilings of disused outhouses (Figures 2.4d, e) with an aspirator (Figure 2.4a). Pootered mosquitoes were transferred to collection cups (Figure 2.4b), labelled with a unique collection number. In addition, the propane-powered Mosquito Magnet® Liberty Pro (Pennsylvania, USA; referred to as Mosquito Magnet Trap ®) trap was used to collect host seeking adults (Figure 2.4c).

Collections were killed either by placing the cup containing the mosquitoes directly into the freezer for 40 minutes or by placing the cup or Magnet net into a sealed plastic bag, containing tissue paper saturated with ethyl acetate. Once the mosquitoes were killed, they were visually sorted to genera and minimums of 3-5 representatives, per species per 41 collection, were pinned for morphology (see section 2.2.1.2). All other mosquitoes were preserved for DNA studies by individually placing in Beem® capsules with their unique labels, before closing. A mounted needle was used to pierce a hole in the capsule to allow escape of moisture. Beem® capsules were stored sealed in plastic bags containing sachets of silica gel at room temperature prior to DNA extraction.

42

Figure 2.4: Adult collection methods and resting habitats: (a) aspiration of adults resting on walls of a Brick Shelter in Kent, (b) collection cups used to transport collected resting adults, (c) Mosquito Magnet trap ® uses carbon dioxide, moisture and heat (bi-products of propane) and a chemical lure (1-Octen-3-ol) to attract host- seeking females (picture from http://amazon.com) (d) a brick shelter in Exminster Nature Reserve, Devon and (e) goat stable in Pettits Animal Farm, Norfolk.

43

2.2.3 Mosquito identification

Using a stereoscope and the morphological identification keys of Cranston et al. (1987) and Snow (1990), both link-reared and resting adult mosquitoes were identified to species in the adult stage, where possible. During the process of collection and transportation, some adult specimens were so badly damaged that species identification was impossible. In addition, not all larvae/pupae collected were successfully reared to adults, these immatures were not identified to species in this study. In such cases, specimens were identified to the genus level only.

A total of 1,463 adult mosquitoes and 1,601 immature stages were collected from over 73 unique sites (Table 2.2) in this study. Species of the Maculipennis Complex were used in the development of an ITS2 PCR-RFLP assay described in Chapter 3 and specimens were then identified as An. atroparvus, An. daciae and An. messeae. Culex pipiens s.l. specimens were used to determine the accuracy of two published assays developed to differentiate between the two forms (Chapter 4).

44

Chapter 3

Molecular differentiation of the Maculipennis Group in the UK

.

45

3. Molecular differentiation of the Maculipennis Group in the UK

3.1 Introduction

3.1.1 The taxonomic status of the Palaearctic Maculipennis Group

Anopheles maculipennis was exposed as the first mosquito sibling species complex, comprising several species on the basis of egg morphology (Falleroni, 1926; van Thiel, 1927; Falleroni, 1932; Corradetti, 1934; de Buck & Swellengrebel, 1934a; Hackett & Lewis, 1935; Weyer, 1942; Angelucci, 1955; Gutsevich et al., 1974; White, 1978; Korvenkontio et al., 1979; Pichot & Deruaz, 1981; Jaenson et al., 1986a; Jetten & Takken, 1994; Sedaghat et al., 2003a; Nicolescu et al., 2004). Despite several other techniques that were later employed to differentiate the component species, including hybridisation experiments (de Buck & Swellengrebel, 1934b; Kitzmiller et al., 1967), detailed morphology (La Face, 1931; de Buck et al., 1933; Diemer, 1935; Bates, 1939; Buonomini, 1940; Ungureanu & Shute, 1947; Işfan, 1952; Suzzoni-Blatger & Sevin, 1981; Boccolini et al., 1986; Suzzoni-Blatger et al., 1990; Deruaz et al., 1991), ecology (van Thiel, 1927; de Buck & Swellengrebel, 1934b; Hackett & Missiroli, 1935), cytotaxonomy (Frizzi, 1952; Frizzi, 1953; Kitzmiller et al., 1967; Stegnii, 1976; Stegnii & Kabanova, 1976; White, 1978), zymotaxonomy (Korvenkontio et al., 1979; Bullini et al., 1980; Bullini & Coluzzi, 1982; Jaenson et al., 1986a; Cianchi et al., 1987; Suzzoni-Blatger et al., 1990) and cuticular hydrocarbons (Phillips et al., 1990), egg morphology remained the golden standard for differentiating species within the complex. However several authors have pointed out that intraspecific variation in egg morphology can result in incorrect identifications (Guy et al., 1976b; Jaenson et al., 1986a; Alten et al., 2000; Linton et al., 2002b).

The relatively recent application of DNA sequence analysis to the Maculipennis Group has proven to be the most reliable method of differentiating the component taxa (Marinucci et al., 1999; Proft et al., 1999; Romi et al., 2000; Linton et al., 2001a; Linton et al., 2002a,b, c; Linton, 2004; Sedaghat et al., 2003a,b; Nicolescu et al., 2004; Gordeev et al., 2004, Gordeev et al., 2005; Linton et al., 2007; Djadid et al., 2007). Indeed DNA sequences were used to prove the synonymy of An. subalpinus with An. melanoon (Linton et al., 2002b) and have revealed three new taxa in the Maculipennis Complex: An. persiensis Linton, Sedaghat &

46

Harbach (Sedaghat et al., 2003b), An. daciae Linton, Nicolescu & Harbach (Nicolescu et al., 2004) and An. artemievi Gordeev, Zvantsov, Goriacheva, Shaikevich & Ezhov (Gordeev et al., 2005).

The Palaearctic Maculipennis Group currently comprises eleven formally recognised species: Anopheles artemievi, An. atroparvus van Thiel, An. beklemishevi Stegnii & Kabanova, An. daciae, An. labranchiae Falleroni, An. maculipennis s.s, Meigen, An. martinius Shingarev, An. melanoon Hackett, An. messeae Falleroni, An. persiensis and An. sacharovi Favre (White, 1978; Linton et al., 2002b; Sedaghat et al., 2003b; Nicolescu et al., 2004; Gordeev et al., 2005).

3.1.2 Molecular differentiation of species in the Maculipennis Group

Molecular methods using the sequence of the second nuclear internal transcribed spacer (ITS2) region of the ribosomal DNA (rDNA) have been widely employed to identify the isomorphic Palaearctic members (Marinucci et al., 1999; Proft et al., 1999; Linton et al., 2001a; 2002a,b,c; Sedaghat et al., 2003a,b; Nicolescu et al., 2004; Gordeev et al., 2004; Gordeev et al., 2005; Kampen, 2005a,b; Linton, 2004; Linton et al., 2005; Linton et al., 2007; Djadid et al., 2007) and to investigate the internal phylogenetic relationships within the Maculipennis Group (Marinucci et al., 1999; Linton, 2004; Kampen, 2005b; Djadid et al., 2007). The high inter-specific divergence of the ITS2 region makes it a useful marker in accurately identifying the members of species complexes (Section 2.1.1).

Proft et al. (1999) developed an ITS2-PCR assay that could differentiate six members of the Maculipennis Group (An. atroparvus, An. labranchiae, An. maculipennis, An. melanoon, An. messeae and An. sacharovi) using species-specific primers. Recently Kampen (2005a) incorporated another species-specific primer for An. beklemishevi into the original assay. Linton et al. (2005) reported that the purported messeae-specific primer could also amplify An. daciae, a recently described member in the Maculipennis Group (Nicolescu et al., 2004). Anopheles daciae and An. messeae are the two most closely related sister taxa within the Maculipennis Group, sharing 99.0% sequence homology in the 485bp ITS2 amplicon, with five fixed variable sites (Linton, 2004; Nicolescu et al., 2004; Linton et al., 2005). In Romania, where An. daciae was originally described, the two were often found in sympatry

47

(Nicolescu et al., 2004). Correlation of mitochondrial sequence data from the cytochrome oxidase I (COI) gene shows the purported presence of An. daciae in Italy, The Netherlands, Former Yugoslavia and Kazakhstan (as An. messeae, Di Luca et al., 2004), suggesting that its Eurasian distribution is extensive (Linton et al., 2005). With the exception of DNA assays, there are currently no other reliable means of differentiating these two species.

3.1.3 Maculipennis Group and malaria transmission

Interest in the Maculipennis Group has been sustained due to their role in malaria transmission in Europe and the Middle East. Three species of the Maculipennis Group, An. atroparvus, An. sacharovi and An. labranchiae, are known to be efficient current or historical malaria vectors in Europe (Hackett & Missiroli, 1935; Bruce-Chwatt & de Zulueta, 1980; Jaenson et al., 1986a; Ribeiro et al., 1988; Kasap, 1990; Jetten & Takken, 1994; Fantini, 1994; Romi et al., 1997; Romi, 1999; Alten et al., 2000; Romi et al., 2001; Romi et al., 2002; Sedaghat et al., 2003a). Anopheles maculipennis s.s. and An. melanoon (as An. subalpinus) were recently incriminated as secondary vectors in the Biga Plains of Turkey (Alten et al., 2000), perhaps indicating an increased role in malaria transmission. There is still debate over the role of Anopheles messeae as a vector. It is reported to be an efficient vector in western Asia (Bruce-Chwatt & de Zulueta, 1980), Ukraine and Russia (Nikolaeva, 1996) but in Europe.

Despite the eradication of malaria from Europe following WWII (Ramsdale & Gunn, 2005), increasing numbers of malaria cases are now being reported (Sartori et al., 1989; Nikolaeva, 1996; Baldari et al., 1998; Romi et al., 2001) heightening concern for the reintroduction of malaria in regions, such as the UK where competent mosquito vectors still exist (Jetten et al., 1996; Romi et al., 1997; Lindsay & Birley, 1996; Romi, 1999; Snow, 1999; Romi et al., 2001; Linton et al., 2001b).

3.1.4 The Maculipennis Group in the UK

Based on egg morphology, Edwards (1936) suggested the presence of three species of the Maculipennis Group in the UK: An. atroparvus, An. messeae and An. maculipennis s.s. However, subsequent morphological studies confirmed the presence of only two species,

48 namely An. atroparvus and An. messeae (Marshall, 1938; Mattingly, 1950; Wallace, 1958; Cranston et al., 1987; Snow, 1990; Snow et al., 1998). Despite the presence of An. maculipennis s.s. in The Netherlands, France, Germany and Belgium, it has not been confirmed in the UK (Cranston et al., 1987; Snow & Ramsdale, 1999; Ramsdale & Snow, 2000), despite a tentative report from the Channel Islands (Ramsdale & Wilkes, 1985 , cited in Cranston et al., 1987). Although An. maculipennis s.l. has been reported from as far north as the Grampians in Scotland (Ashworth, 1927) and in Ireland (Ashe et al., 1991), no formal studies have been carried out to determine which species are actually present (Rees & Snow, 1990). The presence of An. messeae, however, was confirmed in County Galway in Ireland (F. Geraghty & Y.-M. Linton, pers comm), using ITS2 DNA sequence data. Employing the same nuclear gene region, Linton et al. (2002a) verified the presence of both An. atroparvus and An. messeae in Kent and An. messeae in Yorkshire.

Linton et al. (2005) also confirmed the presence of a third member of the Maculipennis Group in the Somerset Levels, UK, Anopheles daciae, through the comparison of ITS2 sequence data from five adults (collected resting in a horse stable in Godney Farm, Godney), with those from the type locality of Budeni, Romania (Nicolescu et al., 2004).

3.1.4.1 Distribution and ecology of the Maculipennis Group in the UK

Anopheles messeae is widely reported in Europe, occurring also in northern China, Mongolia and former USSR states (WRBU, 2008). In the UK it is reported from Wales, England and as far north as central Scotland (Walter, 1927 in Morgan, 1978; Wallace, 1958; Cranston et al., 1987; Rees & Snow, 1990; Snow et al., 1998; Ramsdale & Snow, 2000). The species has been reported in the London area (Epsom, Esher, Dartford, Bexley, Romford, Richmond, Wimbledon Common, Putney and Barking (Nye, 1955) and from the English counties of Berkshire, Cambridgeshire, Cheshire, Devon, Kent, Norfolk, Northumberland, Suffolk, Surrey, Sussex and Yorkshire (Evans, 1934; Carter, 1978; Cranston et al., 1987; Ramsdale & Snow, 2000; Linton et al., direct submission to GenBank, 2001; Linton et al., 2002a).

Anopheles atroparvus has been recorded from Berkshire, Cheshire, Dartford, Devon, Dorset, Essex rivers, Hayling Island, Kent, Pevensey Levels, Romney Marsh, Surrey, Thames

49

Estuary and in the lower reaches of Sussex rivers in England (James, 1929; Shute, 1933; Marshall & Staley, 1933; Killington, 1946; Ramsdale & Snow, 2000; Linton et al., 2002a) and around the London area (Bexley, Dartford, Epsom, Esher, Epping & Romford (Nye, 1955). In northern Wales, An. atroparvus has been recorded from Anglesey, Llanfaglan, the southern coast of the Menai Straits and at Gwyrfai on the River Afon estuary (Wright, 1924; Evans, 1934; Ramsdale & Snow, 2000).

Immatures of An. messeae can be found in inland fresh water pools, such as ponds and ditches, which are either slow moving or stagnant (Cranston et al., 1987; Snow, 1990). In contrast, An. atroparvus immatures are tolerant to high levels of salinity (Wallace, 1958), thus it is predominantly found in brackish open water or weedy ditches in coastal or estuarine locations (Marshall, 1938; Wallace, 1958; Rees & Snow, 1990; Snow, 1990). Although An. atroparvus has also been reported from fresh water (Rodhain & van Hoof, 1942) and has an extensive inland distribution in the Iberian Peninsula (Romeo Viamonte, 1950).

Both An. messeae and An. atroparvus overwinter as nulliparous females (Cranston et al., 1987). Adults of An. messeae are commonly found resting in animal shelters, such as horse stables and abandoned outhouses (Linton et al., 2002a). In winter, they enter complete hibernation, seeking refuge in cold, uninhabited shelters (Marshall, 1938; Cranston et al., 1987) and surviving off their fat body reserves. Capable of entering full diapause (Mohrig, 1969), An. atroparvus tends to hibernate in warmer sheltered spots such as sheds and stables and will periodically break out of hibernation to feed (Cranston et al., 1987; Rees & Snow, 1990).

Anopheles daciae has only been reported from Godney, in the Somerset Levels, to date (Linton et al., 2005). Five females were collected as resting adults within a horse stable, thus apart from adult resting habitat, little is known of the specific ecology of An. daciae in the UK (Linton et al., 2005). However, in the type description, Nicolescu et al. (2004) did report sympatric larval collections of An. messeae and An. daciae in several localities in Romania, suggesting that the habitat requirements for An. daciae and An. messeae immatures could be similar.

50

3.1.4.2 Role of Maculipennis Group in malaria transmission in the UK

Early surveys of mosquitoes in the UK (Nutall et al., 1901; Lang, 1918) showed that the distribution of Anopheles mosquitoes was much more extensive than the regions affected by malaria (Cranston et al., 1987). However, a review of historical documents detailing malaria cases in England and Wales from 1840 to 1910 by Kuhn et al. (2003) showed that although malaria was documented across southern Britain, the highest incidence of malaria was documented in the inland county of Cambridgeshire and the coastal county of Kent where up to 96-114 cases per 100,000 inhabitants were reported annually (Figure 3.1).

51

Figure 3.1 Distribution of benign tertiary malaria in England from 1840 to 1910 (Kuhn et al., 2003). Intensity of colour signifies the annual number of cases per 100,000 inhabitants by county: Maroon =96-114, bright red =51- 65, salmon pink =36-50 and light pink =9-20.

52

Post World War I, the transmission of Plasmodium vivax malaria from soldiers quarantined on the Isle of Grain and Sheppey to the locals in Kent inextricably linked An. atroparvus to the transmission of malaria (Shute, 1963). Although An. atroparvus was regarded as the major vector of P. vivax in the UK (Shute & Maryon, 1974; reviewed in Cranston et al., 1987 & Rees & Snow, 1990), its propensity for coastal sites and the intensity of malaria in other more inland localities (Figure 3.1), suggests that other native species must also have played a role.

Anopheles plumbeus, a tree-hole breeder, was implicated in both an outbreak of P. falciparum in a “northern health resort” (Blacklock & Carter, 1920) and in a P. vivax malaria outbreak in Lambeth, London (Shute, 1954). The efficacy of this species as a vector of Plasmodium was later verified in the laboratory (Marchant et al., 1998; Eling et al., 2003). Although An. plumbeus is highly anthropophilic (Cranston et al., 1987), it seems unlikely that it played a major role in the UK malaria transmission, except in localized areas.

To date, Anopheles messeae is not considered a vector of malaria transmission in the UK (Snow, 1998; Marchant et al., 1998). Its occurrence in Kent (Linton et al., 2005), where the burden of malaria was high and its susceptibility to P. vivax infection (Curtis & White, 1984) was evident, presented itself as a potential vector in the UK. It is thought to be a highly competent vector in Russia and the Ukraine, Eastern Europe and western Asia (Detinova, 1953; Bruce-Chwatt & de Zulueta, 1980; Nikolaeva, 1996), yet it is not considered an efficient vector in northwestern Europe (Jetten & Takken, 1994). Linton et al. (2005) suggested that this conflict may in part be due to the discovery of the sympatric and closely related species, An. daciae, whose true distribution could be masked by An. messeae. They proposed that if only one of these isomorphic taxa were to be a good vector of P. vivax, then this could account for the patchy endemic malaria in regions where only An. messeae has been reported and indirectly incriminated (Linton et al., 2005). Anopheles daciae was recently circumstantially incriminated, as it was the only species collected in close proximity to an indigenous malaria case in southern Romania (Vladimirescu et al., 2006). Given the genetic similarity and reported geographical sympatry of An. messeae and An. daciae and the closely related An. atroparvus, it is important to obtain a robust identification method that will allow for future study of these species to determine vector capacity and define effective control programs both in the UK and across Europe. 53

3.2 Aims

The aims of this study were:

[1] To carry out field collections of the Maculipennis Group in five counties in southern England (Devon, Kent, Norfolk, Somerset and Suffolk) and in Anglesey, North Wales,

[2] To design a molecular assay to reliably differentiate between the three British members of the Maculipennis Group and

[3] To use molecularly identified specimens to determine the current presence of An. atroparvus, An. daciae and An. messeae in six regions of southern England and Wales.

3.3 Materials and Methods

3.3.1 Collection and identification of field-caught specimens

Adults and immatures of An. maculipennis s.l. were collected in various sites across southern England and Wales in July 2006: Devon, Somerset, Norfolk, Suffolk and on the Welsh island of Anglesey and in August 2006, in Kent (Table 3.1). Specimens of the Maculipennis Group were collected as immatures in four sites - in the Exminster Marshes (Devon), in a fenland near Reedham (Norfolk), Carlton Marshes in Oulton (Suffolk) and in the Mallraeth Marshes near Llangristriolus (Anglesey). Resting adults of An. maculipennis s.l. were manually collected from the walls and ceilings of animal stables in Norfolk and Somerset and abandoned pillboxes, brick shelters and war bunkers in Devon, Kent and Somerset (Table 3.1) and processed according to methods outlined in Chapter 2 Section 2.2.

54

County Exact Locality Co-ordinates Date Devon *Stream margin, Exminster 50.6773582'N, 03.4729606'W 06.07.06 Marshes

Brick Shelter 1, Exminster 50.6756321'N, 03.4743221'W 06.07.06 Marshes

Brick Shelter 2, Exminster 50.6752617'N, 03.4751598'W 06.07.06 Marshes

Kent Brick shelter, Cliffe Marshes 51.4537556'N, 00.4782834'E 11.08.06

Bunker, Cliffe Marshes 51.4783052'N, 00.4827834'E 11.08.06

Sheep Corral, Cliffe Marshes 51.4671890'N, 00.4895240'E 11.08.06

Norfolk Pettits Animal Farm, Reindeer 52.5659609'N, 01.5768554'E 20.07.06 and Miniature Horse stables

Pettits Animal Farm, Goat and 52.6614040'N, 01.5768701'E 16.07.06 Rhea Stables

*Stream margin Horsey Road, 52.7215319'N, 01.6550687'E 16.07.06 Reedham

Somerset Pillbox 1, Godney 51.1822719'N, 02.7283164'W 07.07.06

Pillbox 2, Godney 51.1816487'N, 02.7273050'W 07.07.06

Horse stable, Godney 51.1817652'N, 02.7230142'W 07.07.06

Suffolk *Carlton Marshes, Oulton 52.4683907'N, 01.6928074'E 19.07.06

Anglesey *Mallraeth Marsh, 53.2017891'N, 04.3454238'W 14.07.06 Llangristriolus C

Table 3.1 Dates and exact collection localities of An. maculipennis s.l. collected across southern England and Wales in June and August 2006. *Indicates immature collections, link-reared to adults; all others were collected as resting adults.

55

Specimens were morphologically identified as belonging to An. maculipennis s.l., primarily by their distinctive spotted wings, using the morphological key of Snow (1990), then identified to species either by direct sequencing of the nuclear ribosomal ITS2 (second internal transcribed spacer) region (see section 3.3.3), or using the ITS2 PCR-RFLP assay designed herein (see section 3.3.4). Voucher specimens (morphological specimens and/or DNA extractions) are available in the Natural History Museum, London, for future reference.

3.3.2 DNA extraction and ITS2 PCR amplification

DNA was extracted from individual specimens following the phenol-chloroform extraction protocol of Linton et al. (2001b). PCR amplification of the ITS2 region was achieved using the 5.8SF and 28SR primers of Collins & Paskewitz (1996). PCR reaction mixes and thermocycler parameters used in this study were those previously developed and detailed in Linton et al. (2001b). PCR products were amplified using either 2µl of template DNA or using a single mosquito leg placed directly in the PCR mix (Scott et al., 1993). PCR products destined for sequencing were first cleaned using the commercially available QIAgen PCR Purification Kit (QIAgen Ltd, Sussex, England), following the manufacturers instructions.

3.3.3 Direct sequencing of the ITS2 amplicon

Amplicons of ITS2 were directly sequenced for specimens of An. maculipennis s.l. collected in Norfolk (n=77), Devon (n=21), Kent (n=9) and Somerset (n=10). Purified ITS2 PCR products were sent to the Zoological Sequencing Facility in the Natural History Museum for sequencing. Resultant sequences were edited and aligned using SequencherTM version 4.6 (Genes Codes Corporation, Ann Arbor, Michigan) and CLUSTAL W (http://align.genome.jp/, Thompson et al., 1997). The FASTA search engine (http://www.ebi.ac.uk/fasta33/) was used to assess the similarity of ITS2 sequences generated in this study with those in GenBank. ITS2 sequences generated in this study were identified by comparison with those available in GenBank (An. atroparvus from England and Romania (487 bp; Linton et al., 2002a; Nicolescu et al., 2004), An. daciae from England and Romania (485bp; Linton, 2004; Nicolescu et al., 2004; Linton et al.,

56

Species Country Locality GenBank References accession Numbers An. England Kent (12) AF504237-248 Linton et al., 2002a atroparvus (12) (42)

Romania Budeni (18) AY634505-522 Nicolescu et al., 2004 (30) Saftica (12) AY634523-534 Nicolescu et al., 2004

An. England Somerset (5) AY822585-589; Linton et al., 2005 daciae (5) (103)

Romania Budeni (63) AY634407-469 Nicolescu et al., 2004 (98) Constanta (13) AY634470-482 Nicolescu et al., 2004 Giurgiu (1) AY634406 Nicolescu et al., 2004 Mehedinti (2) AY634483-484 Nicolescu et al., 2004 Saftica (19) AY634485-503 Nicolescu et al., 2004

An. England Kent (40) AF504197-236 Linton et al., 2002a messeae (43) London (1) AF504196 Linton, dir. sub. 2001 (65) Yorkshire (2) AF452699-700 Linton, dir. sub. 2001

Greece Florina (2) AF342711-712 Linton et al., 2002c (2)

Romania Budeni (1) AY648982 Nicolescu et al., 2004 (17) Mehedinti (16) AY648984-998 Nicolescu et al., 2004 EF090197 Nicolescu et al., 2004

Sweden Moja Island (3) EF090194-196 Linton, dir. sub. 2001 (3)

Table 3.2 GenBank accession numbers and locality of 210 ITS2 sequences [An. atroparvus from England and Romania (n=42), An. daciae from England and Romania (n=103) and An. messeae from England, Greece, Romania and Sweden (n=65)] used to design the ITS2 PCR-RFLP assay designed to differentiate the three members of the British Maculipennis Group together with sequences generated in this study.

57

2005) and An. messeae from England, Greece, Romania and Sweden (485 bp; Linton et al., direct submission 2001; Linton et al., 2002a,c; Nicolescu et al., 2004) (Table 3.2).

3.3.4 Development of ITS2 PCR-RFLP assay

The ITS2 sequences (n=117) generated in this study (Table 3.3) were aligned together with the 210 available ITS2 sequences of An. atroparvus (487bp), An. daciae (485bp) and An. messeae (485bp) in GenBank (Table 3.2). Consensus sequences of An. atroparvus, An. daciae and An. messeae were aligned using Clustal W and used to identify appropriate fixed sequence differences within the fragments which could be exploited by restriction enzymes to differentiate between the three members of the British Maculipennis Group. The computer software Mapper, available online at http://arbl.cvmbs.colostate.edu/molkit/mapper/index.html, was used to determine enzyme choice. The optimal enzyme BstU I (CG↓CG) was chosen as it resulted in sufficiently different sized fragments that were species-diagnostic for An. atroparvus (445 & 42 bp), An. daciae (332, 59, 52 & 42bp) and An. messeae (332, 109 & 42bp). Restriction sites for each species are indicated in Figure 3.3 and resultant fragment sizes following electrophoresis are shown in Figure 3.2.

RFLP digestions were carried out in 20µl reactions as follows: 4 µl cleaned ITS2 PCR product, 13µl ddH2O, 2µl Buffer 2* and 1µl BstU I enzyme* (*New England BioLabs). The reactions were incubated at the optimal enzyme activity temperature of 60°C in a thermocycler for a minimum of 3 hours to ensure full digestion of the fragments. Restriction fragments were visualised following electrophoresis on a 3% agarose gel containing 1% ethidium bromide for 1 hour at 70V. Fragment sizes were measured using Hyperladder IV (BioLine) (Figure 3.2).

58

1 2 3 4 5 6 7

An. atroparvus An. messeae An. daciae

Figure 3.2 Species-diagnostic ITS2 products from the three British Maculipennis Group species following digestion with BstU I enzyme (CG↓CG). Lane 1: Hyperladder I 100 bp ladder (BioLine). Lanes 2 & 3: An. atroparvus (445 bp, 42 bp). Lanes 4 & 5: An. messeae (332bp, 109bp, 42bp). Lanes 6 & 7: An. daciae (332bp, 59bp, 52bp, 42bp). Note: fragments under 100bp are generally not visible on this agarose gel.

59

3.4 Results

A total of 711 specimens of the Maculipennis Complex collected across southern England and northern Wales were molecularly identified to species in this study (Table 3.3). One hundred and seventeen nuclear DNA (nDNA) ITS2 sequences were generated (Table 3.3) and compared with reference sequences from England, Greece, Romania and Sweden available in Genbank (Table 3.2). The remaining 594 samples were identified using the ITS2 PCR-RFLP designed in this study (Table 3.3).

Of the counties sampled all three members were present in two – Kent and Norfolk. In Somerset and Suffolk, An. messeae and An. daciae were collected in sympatry, as adults and as larvae, respectively. In Somerset and Anglesey, only An. daciae was detected, while only An. messeae were collected in Devon.

3.4.1 ITS2 sequences

No intra-specific variation was noted within the ITS2 sequences of An. atroparvus, An. daciae and An. messeae generated from UK populations and were identical to those previously sequenced from England, Greece, Romania and Sweden (Table 3.2). Anopheles daciae and An. messeae share 99% ITS2 sequence identity, with only five fixed species-diagnostic sites at bases 214, 218, 220, 416 and 436 of the alignment with An. atroparvus (Figure 3.3). The ITS2 sequence of An. atroparvus was more divergent, with 46 divergent bases (90.6% identity) and 47 (90.4% identity) fixed nucleotide differences from An. daciae and An. messeae, respectively (Figure 3.3).

60

County Exact localities An. An. An. atroparvus daciae messeae (n=111) (n=471) (n=129) Anglesey (n=1) Mallraeth Marshes, Llangristriolus C. 0 1 0 Devon (n=99) *Exminster Marshes (Stream margin) 0 0 2 Exminster Marshes (Brick shelter 1) 0 0 71(19) Exminster Marshes (Brick shelter 2) 0 0 26(2) Kent (n=110) Cliffe marshes, Cliffe (Brick shelter 1) 5 0 2 Cliff marshes, Cliffe (Salt Lane 4 0 2 Bunker) Cliffe marshes (Sheep corral) 88(3) 6(6) 3 Norfolk (n=402) Goat Stables, Pettits Animal Farm 8(5) 198(43) 2(1) Reindeer and miniature Horse stables, 6(1) 175(27) 11 Pettits Animal Farm Stream on Horsey Road, Reedham 0 2 0 Somerset (n=93) Godney Village (Pillbox 1) 0 63(1) 5 Godney Village (Pillbox 2) 0 15 1 Godney Farm, Godney (Horse stable) 0 8(8) 1(1) Suffolk (n=6) *Carlton Marshes, Oulton 0 3 3

Table 3.3 Relative proportions of species of the Maculipennis Group (n=711) collected and molecularly identified in five English counties (Devon (n=99), Norfolk (n=402), Somerset (n=93), Suffolk (n=6) and Kent (n=110) and on the Welsh island of Anglesey (n=1) in this study. Anopheles atroparvus (n=111), An. daciae (n=471) and An. messeae (n=129) Total numbers identified by the ITS2 PCR-RFLP assay and by direct sequencing of the ITS2 fragment (numbers sequenced are shown in parenthesis) are shown. *indicates larval collections.

61

3.4.2 PCR-RFLP Assay

All previously sequenced (n=210) and newly generated ITS2 sequences (n=117) from a wide range of geographical locations (Table 3.3) were aligned using Clustal W (http://align.genome.jp/). As no intraspecific variation was noted, irrespective of geographical origin of the samples, a representative ITS2 sequence for each of An. atroparvus, An. daciae and An. messeae were aligned. Polymorphic portions of the alignment were identified and screened for potential cutting sites using restriction enzymes (Figure 3.3).

In addition to the 117 specimens of the Maculipennis Group identified by ITS2 sequencing, a further 594 were identified following screening with the ITS2 PCR-RFLP assay designed herein (Table 3.3). Of the 711 specimens identified in this study, 11 individuals identified were collected as immatures, with the remaining 700 individuals were collected as resting adults. Anopheles daciae comprised 65.6% of the total number sampled, 17.9% were identified as An. messeae and 15.5% were An. atroparvus (Table 3.3).

62

1 1111111112 2222222223 3333333334 4444444445 5555555556 6666666667 7777777778 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 messeae atcactcggc tcgtggatcg atgaagaccg cagctaaatg cgcgtcacaa tgtgaactgc aggacacatg aacaccgata daciae ...... atroparvus ......

1 1111111111 1111111111 1111111111 1111111111 1111111111 1111111111 8888888889 9999999990 0000000001 1111111112 2222222223 3333333334 4444444445 5555555556 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 messeae agttgaacgc atattgcgca tcgtgcgaca cagctcgatg tacacatttt tgagtgccca tatttgaccc --attcaagt daciae ...... --...... atroparvus ...... t.a ta.cc....c

1111111111 1111111111 1111111111 1111111112 2222222222 2222222222 2222222222 2222222222 6666666667 7777777778 8888888889 9999999990 0000000001 1111111112 2222222223 3333333334 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 messeae caaactacgt acctccgtgt acgtgcatga tgatgaaaga gtttgga-ac accttccttc -tcttgcatt gaaagcgcag #daciae ...... -.. ...a...a.t -...... #atroparvus.....gg...... acc...... g.-.t ...... g ..c....t.. g..a.....t c...... c ...gt..t..

2222222222 2222222222 2222222222 2222222222 2222222222 2222222223 3333333333 3333333333 4444444445 5555555556 6666666667 7777777778 8888888889 9999999990 0000000001 1111111112 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 #messeae cgtgtagcaa ccccaggttt caacttgcaa agtggccatg gggctgacac ctcaccacca tcagcgtgct gtgtagcgtg #daciae ...... #atroparvus ......

3333333333 3333333333 3333333333 3333333333 3333333333 3333333333 3333333333 3333333334 2222222223 3333333334 4444444445 5555555556 6666666667 7777777778 8888888889 9999999990 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 messeae ttcggcccag taaggtcatc gtgaggcgtc acctaacggg gaagcacaca ctgttgcgcg tatctcgtgg ttctaaccca daciae ...... atroparvus ...... tc...... t ...... a ...... cag ...c....t...... a... ..acc-....

4444444444 4444444444 4444444444 4444444444 4444444444 4444444444 4444444444 4444444444 0000000001 1111111112 2222222223 3333333334 4444444445 5555555556 6666666667 7777777778 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 1234567890 messeae accatagcag cagaggtaca agaccagctc ctagcggcgg gagctcatgg gcctcaaata atgtgtgact accccctaaa daciae ...... a...... c...... atroparvus ...... a...... a...... t...... a......

444444444 888888888 123456789 messeae tttaagcat daciae ...... atroparvus ......

Figure 3.3 The 489bp alignment of nuclear ITS2 sequences of An. messeae (485bp), An. daciae (485bp) and An. atroparvus (487bp). Enzyme BstU I (CG↓CG) cutting sites are highlighted in yellow. One site is present at 42 bp for all three species, a second site is present for both An. messeae and An. daciae at 378bp and a third site is present for An. daciae only at 437bp, thus the following species-diagnostic fragments are generated: An. atroparvus (445 & 42 bp), An. daciae (332, 59, 52 & 42bp) and An. messeae (332, 109 & 42bp).

63

3.4.3 Ecology

Anopheles daciae was collected as immatures (n=6) in Mallraeth Marshes in Anglesey (n=1, pupae), in Carlton Marsh Reserve, Suffolk (larvae, n = 3) and in a ditch in fenland near Reedham, Norfolk (larvae, n=2) and as resting adults in various animal stables (rhea, goat, reindeer, miniature donkey) in Pettits Animal Farm in Reedham, Norfolk (n=373), in Godney farm and Garslade farm in Somerset (n=86) and in the Cliffe marshes, Kent (n=6).

A single collection consisting of An. messeae was collected as immatures (n=5) in a stream margin in Exminster Marshes, Devon (n=2 pupae). Anopheles messeae was also collected in the Carlton Marsh Reserve, Suffolk (n=3; 1 larva, 2 pupae) in sympatry with An. daciae. Adults of An. messeae (n=122) were collected in the Exminster Marshes in Devon (n=97), in adjacent reindeer (n=11) and goat stables (n=2) at Petitts Animal farm in Norfolk, in Godney farm and Garslade farm in Somerset (n=7) and in the Cliffe Marshes in Kent (n=7). The only member of the Maculipennis Group collected as larvae and adults in the Exminster RSPB Nature Reserve, near Exeter, were An. messeae (n=99).

Anopheles atroparvus was only collected as adults in two counties in this study - in reindeer stables in Pettits Animal Farm in Norfolk (n=14) and resting in disused bunkers in the Cliffe Marshes in Kent (n=97), being particularly abundant in one bunker currently sused as a sheep corral.

3.5 Discussion

Historically, identification of British members of the Maculipennis Group relied on the indirect method of comparison of egg morphology or ecological nuances, such as incomplete diapause, to differentiate between An. atroparvus and An. messeae. Linton et al. (2002a) were the first to use ITS2 DNA sequencing to identify An. atroparvus and An. messeae in the Cliffe Marshes in Kent and ITS2 sequencing of specimens of purported “An. messeae” in Somerset, revealed the presence of An. daciae for the first time (Linton et al., 2005). The costs of DNA sequencing is prohibitive when considering a wider study on current distribution of the Maculipennis Group, but herein an accurate, inexpensive method of processing large numbers of specimens is presented. A total of 711 specimens collected in July and August 2006 from 64 northern Wales (Anglesey) and England (Devon, Norfolk, Somerset, Suffolk and Kent) were identified to species in this study, thus comprising the largest species-level study of the British Maculipennis Group to date. Interestingly this study suggested a widespread occurrence of An. daciae, with 65.6% of all specimens collected in southern Britain being of this species.

The most interesting result obtained in this study, aside from the presence of An. daciae in Godney, Somerset, is the documented presence of An. daciae in Norfolk, Somerset, Suffolk and Kent herein all comprise new distribution records and the collection of a single individual in Anglesey is a new country record for Wales. New distribution records were also established for An. atroparvus from Norfolk and An. messeae in Somerset. Adults of An. daciae were collected resting in sympatry with both An. messeae and An. atroparvus in animal stables and disused war installations in Kent and in Norfolk and with An. messeae only in Somerset. As observed in Romania (Nicolescu et al., 2004), adults of An. daciae in the current study were collected in sympatry with An. atroparvus in Kent and Norfolk and immatures of An. daciae were collected in sympatry with An. messeae in Suffolk.

The main purpose of this study was to develop a PCR assay that could differentiate the members of the Maculipennis Group. The most important objective was to sample all genetic variation found within each species for the UK in order to ensure that the assay was based on fixed polymorphisms rather than on nucleotides that are polymorphic within UK populations. A broad geographical sampling of the Maculipennis Group was thus undertaken to ensure sufficient material for the development of a robust assay. Specimens of the Maculipennis Group were collected in 14 localities and collections of both resting adults and larvae were aimed at maximum capture rates. The extent of sampling was not designed to establish accurate distribution records, and because collections were not standardised, detailed conclusions on the species abundance, composition and occurrence within and between sites were not appropriate. For example, the presence of a single An. daciae individual in Anglesey indicates the presence of the species, but does not exclude the possibility of either An. messeae or An. atroparvus being present. To measure species abundance and richness in each site and/or to compare the prevalence of species between sites, standardised collection methods such as a predetermined number of dips per larval site, collecting along a transect in an aquatic environment, or to collect resting adults for a predetermined period of time over a day for a set number of days (for example collecting every 15min 3 times a day for one week) (Service,

65

1998) are required. Data gathered from such collections could then be statistically examined and a quantitative spatial distribution pattern of the Maculipennis Group could be determined.

Prior to this study, ecological studies carried out on the Maculipennis Group did not differentiate between the three members (Snow, 1998; Hutchinson et al., 2007; Snow & Medlock, 2008). As a result, detailed knowledge of the distribution of individual species within the group was undocumented. Results generated from the assay developed here (section 3.3.4) have confirmed the occurrence of these species in southern England, have contributed significantly to the knowledge of resting adult habitats of An. atroparvus, An. daciae and An. messeae, and the assay will enable fast and accurate studies to be carried out in the future. The detection of An. daciae in 3 English counties and in Anglesey implies that its presence was masked by that of An. messeae, as suggested by Linton et al. (2005). It also suggests that An. daciae could be widespread, cryptically co-occurring with the currently proposed global distribution of An. messeae. Linton et al (2005) compared mitochondrial Cytochrome Oxidase I (COI) of An. daciae from the type series (Nicolescu et al., 2004) with the published study of Di Luca et al. (2004) and showed that some purported An. messeae from The Netherlands, Kazakhstan and Italy were in fact An. daciae. This indicates that the species is present in various places across Europe and in the former USSR states; As well as its documented presence in the UK and Romania, An. daciae has also recently been detected in Poland and Bulgaria (Y-M. Linton, Pers. Comm.).

A recent study of ITS2 sequences in six purported An. messeae populations from Russia (Bezzhonova & Goryacheva, 2008) revealed polymorphism in eight sites along the ITS2 region (including the five daciae-specific bases of Nicolescu et al. (2004). Also documented was intragenomic variation, whereby one individual which showed both An. daciae and An. messeae ITS2 haplotypes. Due to this, the authors proposed that ITS2 was not an effective marker to discriminate between An. daciae and An. messeae and proposed the synonymy of An. daciae with An. messeae. I disagree with this for several reasons. Firstly, both species are found in sympatry in several areas of their range (Nicolescu et al., 2004; Linton et al., 2005; herein). Given that An. daciae and An. messeae are the most closely related species in the Maculipennis Group (Linton et al., 2004; Nicolescu et al., 2004; Linton et al., 2005) and that they occur in sympatry (Nicolescu et al., 2004; herein); it is plausible that these two species have recently undergone ecological speciation. Concerted evolution

66 within the ITS2 can result in the homogenisation of polymorphic loci within a species in a fairly short period of time (Collins & Paskewitz, 1996) thus allowing for successful differentiation of species over a geographical range (Fritz et al., 1994; Navajas et al., 1998); the presence of intra-individual / intra-population variation in the nuclear ITS2 gene has been attributed to the intermixing of differentiated ITS2 populations (Vogler & DeSalle, 1994). Intragenomic ITS variation, on both the individual and population level, has been reported in other species complexes, e.g. Australian tiger beetles - Cicindela dorsalis (Vogler & DeSalle, 1994), in the Ixodes ricinus complex (Wesson et al., 1993) and in mosquitoes, Anopheles albitarsis complex (Wilkerson et al., 2005). Despite this variation, the phylogenetic importance of fixed loci supported the recognition of morphologically distinct subspecies (Vogler & De Salle, 1994), as do the five fixed ITS2 differences of Nicolescu et al. (2004). Given the nature of PCR (i.e. amplification of the most prolific haplotype), low level polymorphic intra-individual ITS haplotypes are usually only revealed by cloning, as in the study of Bezzhonova & Goryacheva (2008). However, sequence data of the independent mitochondrial COI gene from individuals identified by their signature ITS2 gene region do corroborate the separate species status of these taxa (Linton, 2004; Nicolescu et al., 2004; Di Luca et al., 2004). Interestingly, Bezzhonova & Goryacheva (2008) do report the same five fixed base differences between An. daciae and An. messeae (Nicolescu et al., 2004; Linton et al., 2005) and the presence of the diagnostic RFLP restriction site (Section 2.3.4). Thus establishing the validity of An. daciae as a species and suggesting its presence in Russia.

The ITS2 PCR-RFLP assay herein was designed specifically for the differentiation of British members of the Maculipennis Group with a particular focus on An. daciae and An. messeae. With that said, identification of these two species can still be carried out using this RFLP outside the UK. However, due to the short fragment size of the ITS2 (circa 470-490 bp in most members of Palaearctic Maculipennis Group, except An. beklemishevi (853bp) (Kampen, 2005b) and the close genetic relationship between the species, other closely related species could produce similarly sized fragments following the restriction with enzyme BstU I (CG↓CG). For example, this assay will not work effectively in The Netherlands as An. maculipennis s.s. is present in addition to the three British species. Digestion of An. maculipennis s.s. ITS2 amplicons (472bp) with this enzyme would yield fragment sizes of 332 & 103 bp which, when visualised on an agarose gel, could be mistaken for An. messeae. Thus, to avoid such misidentification, An. messeae could first be identified, using the PCR assay of

67

Proft et al. (1999), from the rest of the Palaearctic Maculipennis Group using their purported messeae-specific primer. These individuals could then subsequently be differentiated as either An. messeae or An. daciae (both will be identified as An. messeae), using the ITS2 PCR-RFLP described in this chapter.

The spread of emerging and re-emerging diseases such as malaria and WNv into Europe and the impact of global warming have forced scientists to reassess the potential introduction of these diseases into the British Isles (Medlock et al., 2006) and Europe as a whole. Increased global travel results in approximately 2000 cases of malaria imported annually in the UK (HPA, 2008) presenting the possibility of UK mosquitoes becoming infected with imported Plasmodium as previously documented in the north Kent marshes (Shute, 1963) and in two cases of airport malaria reported near Gatwick (Whitfield et al., 1984). The widespread presence of An. daciae and its sympatry with An. atroparvus in Kent and Norfolk detailed in this study presents this species as a potential vector of malaria in the UK.

It is essential that the vector competencies of our endemic species are carefully assessed with wild populations and that the distribution and ecological factors of all British mosquitoes, especially the Maculipennis Group, are documented. This assay allows cheap, accurate identification of members of the Maculipennis Group and provides a solid identification tool for use in such projects in future. Specific host selection of members of the Maculipennis Group is assessed in Chapter 4.

68

Chapter 4

A critical assessment of molecular identification tools for the Palaearctic members of the Pipiens Group (Diptera: Culicidae)

69

4. A critical assessment of molecular identification tools for Palaearctic members of the Pipiens Group (Diptera: Culicidae)

4.1 Introduction

4.1.1 Taxonomic status and distribution of the Pipiens Group

The genus Culex has a worldwide distribution and comprises more than 762 species in 26 subgenera (Mosquito Taxonomic Inventory, 2008). Culex (Culex) pipiens Linneaus is the nominotypical species of the genus Culex and was originally described from Near Lake Krankesjo, Silvakra farm, Veberod, Scania in Sweden by Linnaeus in 1758. The subgenus Culex comprises the Pipiens Group (5 species), the Sitiens Group (6 species) (Edwards, 1932; Harbach, 1988) (see Chapter 1, Figure 1.3), the South Pacific Atriceps Group (3 species) (Belkin, 1962) and the Duttoni Group (Harbach, 1988) for the monotypic Afrotropical Cx. duttoni Theobald.

The Pipiens Group is further divided into the Pipiens, Trifiliatus, Theileri and Univittatus subgroups (Harbach, 1988). The Pipiens subgroup currently comprises Cx. pipiens Linnaeus (with its two forms, pipiens and molestus), Cx. quinquefasciatus Say, Cx. pallens Coquillett, Cx. australicus Dobrotworsky & Drummon and Cx. globocoxitus Dobrotworsky. Of these, three exhibit limited distributions: Cx. pallens (Japan, Korea and Mexico) and Cx. australicus and Cx. globocoxitus (both species in Australia and Oceania) (Smith et al., 2005; WRBU, 2008). Culex pipiens, on the other hand, is practically pan-global (WRBU, 2008). Three species comprise the Trifiliatus subgroup, namely Cx. restuans Theobald (Nearctic region), Cx. torrentium Martini (Palaearctic region) and Cx. vagans Wiedemann (Oriental and Asiatic Regions). Of the members in the Pipiens Group, Cx. pipiens s.l. (Pipiens subgroup) and Cx. torrentium (Trifiliatus subgroup) are present in the UK (Snow, 1990).

70

4.1.2 The Pipiens Complex

Culex pipiens s.l. is found in urban and semi-urban areas across eastern and western Europe, the Middle East and North America (Vinogradova & Fomenko, 1968; Snow, 1990; Spielman, 2001; Shaikevich, 2007; Almeida et al., 2008; WRBU, 2008). Possibly as a result of its wide geographic occurrence, the species currently has 37 valid synonyms (WRBU, 2008).

Based on differences observed in morphology and host selection, Culex molestus Forskål was described as a seperate species from Cx. pipiens by Forskål in 1775 (in Harbach, 1984) in Rosetta, Kahira and Alexandria in Egypt. The inadequate original description and the lack of a type specimen meant that no subsequent studies of the type of Culex molestus could be made, leading to confusion regarding its specific status. In an early published study, Marshall & Staley (1937) regarded Cx. molestus as a seperate species and listed a number of morphological and ecological factors to differentiate these two species. These include average siphonal lengths, the presence of white scales on the legs as well as the length difference between palpal segments and the proboscis in males (Table 4.1). However, in later publications, Cx. molestus was considered to be a biotype of Cx. pipiens and not a seperate species in the Pipiens Complex (Barr, 1957; Stone et al., 1959).

In order to clarify the taxonomic nomeclature and form a solid foundation for future studies on the Pipiens Complex, neotypes were designated for Cx. molestus (Harbach et al., 1984) and Cx. pipiens (Harbach et al., 1985). Given the lack of reliable morphological characters between the two species, Harbach et al. (1984) proposed that Cx. molestus should be regarded as a junior synonym of Cx. pipiens; as a ‘behavioural/physiological variant’ of Culex pipiens rather than a separate species. Following these works, Culex pipiens s.l. is recognised as comprising two biological forms: Culex pipiens f. pipiens and Culex pipiens f. molestus.

Due to the difficulty in identification based on morphology, ecological parameters are most commonly used to differentiate between the two forms of Cx. pipiens. Culex pipiens f. pipiens is reported to be ornithophilic, eurygamous, anautogenous and its larval habitats are above ground, whereas Culex pipiens f. molestus immatures are found in underground sites 71

(flooded basements, underground tunnels). Culex pipiens f. molestus also is autogenous and anthropophilic and the adults are reportedly able to mate in small confined spaces (stenogamous) (Roubaud, 1933; Tate & Vincent, 1936; Mattingly, 1953; Wallace, 1958; Vinogradova & Fomenko, 1968; Bryne & Nichols, 1999). Unfortunately, in comparisons of autogenous and anautogenous populations in Spain (Chevillon et al., 1995), Israel and Egypt (Nudelman et al., 1988), little or no variation of the other behavioural and ecological characteristics (e.g., anthropophily, habitat preference) was observed. In addition, hybridisation of the forms has been reported in populations in Russia (Vinogradova, 1966; Shaikevich, 2007) and in North America (Barr, 1967; Bahnck & Fonseca, 2006), resulting in further ambiguity (Bryne & Nichols, 1999).

72

Species/Characters Culex pipiens f. pipiens Culex pipiens f. molestus Larvae

• Average value 5.0 <4.3 of siphonal index

Adults

• Common Darker brown Lighter colouration characters Whitish scales at the tip Spots at the tips of femora of the femora and tibia and tibia not conspicuous forming conspicuous spots

Median and lateral No such scale patches patches of dark scales present on ventral surface of abdomen

• Male Combined length of Combined length of terminal 4 palpal terminal 4 palpal segments is longer than segments is shorter than the overall length of that length of proboscis proboscis

• Female Pale tergal bands Tergal bands are not constricted laterally and constricted centrally

Table 4.1 Morphological characters distinguishing Culex pipiens f. pipiens Linnaeus and Culex pipiens f. molestus Forskål (after Marshall & Staley, 1937).

73

4.1.3 Differentiation of Cx. pipiens f. pipiens, Cx. pipiens f. molestus and Cx. torrentium

The morphological similarity of Cx. pipiens f. pipiens and Cx. pipiens f. molestus led to the development of molecular tools to distinguish the two forms. Length variation in the dinucleotide (TG-repeat) microsatellite locus CQ11 allowed for the molecular differentiation between Cx. pipiens f. pipiens and Cx. pipiens f. molestus (Bahnck & Fonseca, 2006). A single 180-bp fragment containing 6 TG repeats indicates the presence of Cx. pipiens f. pipiens while a single 250-bp fragment indicates the presence of Cx. pipiens f. molestus in which the CQ11 locus is absent. The occurrence of both fragments was interpreted to indicate hybrids of the two forms (Bahnck & Fonseca, 2006). This molecular tool provides a relatively inexpensive identification method as individuals can be scored directly following electrophoresis of the PCR product.

The co-occurrence of the closely related Cx. torrentium with Cx. pipiens s.l. further complicates correct identification of Pipiens Group members in Europe. This common and often sympatric mosquito differs from Cx. pipiens s.l. only in the phallosomic structure on the male hypopygium (Martini, 1925; Mattingly, 1952; Service, 1968a) and the presence of a patch of prealar scales in the females (Jupp, 1979; Harbach et al., 1985).

Given the relative difficulty in morphologically differentiating these species in Europe, Shaikevich (2007) developed a two-step RFLP assay based on fixed differences in the mtDNA COI gene between Cx. pipiens f. pipiens, Cx. pipiens, f. molestus and Cx. torrentium (see Figure 4.1a-c). The enzyme Hae III (GG↓CC) cleaves Culex pipiens f. pipiens at the site indicated in Figure 4.1a, discriminating it from both Cx. pipiens f. molestus and Cx. torrentium. The second digest, using the enzyme Bc II (T↓GATCA), cuts Cx. torrentium at only one site (79th base, Figure 4.1b) but cuts Cx. pipiens f. pipiens and Cx. pipiens f. molestus (Shaikevich, 2007) at two recognition sites (Figure 4.1b & c) at bases 79 (Figure 4.1b) and 485 (Figure 4.1c). Thus facilitating the identification of these species.

74

Figure 4.1a

Figure 4.1b

Figure 4.1c

Figure 4.1: Alignment of Cx. pipiens f. pipiens, Cx. pipiens f. molestus and Cx. torrentium using Mesquite (v2.6) showing the three restriction sites in the 710-bp fragment of the COI gene used by Shaikevich (2007). Figure 4.1(a) shows a single A-G polymorphism at base 205 that differentiates Cx. pipiens f. pipiens from Cx. pipiens f. molestus and Cx. torrentium. This difference is exploited in the first RFLP assay using enzyme Hae III (GG↓CC), which cuts the Cx. pipiens f. pipiens fragment into two, but leaves the Cx. pipiens f. molestus and Cx. torrentium uncut. Figures 4.1 (b & c) show the restriction sites of the second enzyme, Bc II (T↓GATCA), which cleaves Cx. torrentium at only one site (79bp) (Figure 3.3b) while digesting Cx. pipiens f. pipiens and Cx. pipiens f. molestus at two sites (79 and 485bp) (Figure 4.1.b, c).

75

4.1.4 Vector status of Culex pipiens s.l.

Culex pipiens s.l. has been incriminated as a vector of several flaviviruses world-wide: Japanese Encephalitis in Southeast Asia and Australia (Johansen et al., 2002), Ockelbo virus in Sweden (Jaenson et al., 1986b; Lundström et al., 1990b), Rift Valley Fever in Egypt (Meeghan et al., 1980; Turrell et al., 1996), Saint Louis Encephalitis in North America (Meyer et al., 1982; Miller et al., 1996) and West Nile virus (WNv) in Europe and North America (Hayes, 2001; Fonseca et al., 2004; Hamer et al., 2008; see Chapter 1, section 1.4.2). Although (WNv) was first reported in the Rhone delta in 1963, the recent global resurgence of the virus in southern France (Balenghien et al., 2006), Italy (Romi et al., 2004, in horses), Portugal (Estevez et al., 2005; Almeida et al., 2008), Romania (Tsai et al., 1998; Savage et al., 1999), Russia (Lvov et al., 2000; Platonov et al., 2001) and several states in North America (Marfin et al., 2001) has rejuvenated research efforts in identifying mosquito vectors and routes of transmission. As well as circumstantial association in WNv endemic areas via patient screening and collections of dead birds (Marfin et al., 2001), the ubiquitous Culex pipiens s.l. was incriminated as an efficient vector of WNv following direct isolation of the virus from the mosquito (Romi et al., 2004; Almeida et al., 2008; Hamer et al., 2008), host selection studies (Balenghien et al., 2006; Hamer et al., 2008) and laboratory infection tolerances (Lundström et al., 1990).

In the UK, Buckley et al. (2003) provided evidence of WNv, Sindbis virus and Usutu infections in native British birds by screening 353 serum samples for the presence of antibodies to these viruses. Fifty-two (14.7%) of the birds tested were positive for WNv, while two samples each were positive for Usutu and Sindbis, respectively. This was the first indication of the presence of WNv in the British Isles. A later study of sentinel chickens found that 46-day old chicks had neutralising antibodies to WNv, proving the antibodies were not maternally transferred; implying that active transmission of the virus was occurring between endemic birds and mosquitoes in the UK (Buckley et al., 2006). The detection of WNv in local birds is of public health concern, especially given the reported human cases of the virus in the USA (Lanicotti et al., 1999) and in Europe (Almeida et al., 2008). The source of the virus is most likely the constant influx of birds stopping off in the UK on migratory routes from Africa (Rappole et al., 2000; Higgs et al., 2004; Medlock et al., 2005). Interestingly, the

76 ornithophilic Cx. torrentium is reported as a vector of avian Sindbis virus (Medlock et al., 2007), which has been detected in endemic UK birds (Buckley et al., 2003). Its role in WNv transmission remains unclear.

Given the established importance of Cx. pipiens s.l. and Cx. torrentium in arbovirus transmission (Lundström et al., 2000; Medlock et al., 2007), it is essential to be able to accurately identify the exact composition and distribution of these taxa in the UK as the basis for a strategic action plan should these diseases become a problem in the UK in the future.

4.2 Aims

The aims of this study were:

[1] To assess the presence of Cx. pipiens f. pipiens and Cx. pipiens f. molestus in the UK using the CQ11 microsatellite assay of Bahnck & Fonseca (2006) and

[2] To test the congruence of the results of the CQ11 assay with the mtDNA COI assay of Shaikevich (2007).

4.3 Materials and Methods

4.3.1 Collection of Culex

Immature collections were carried out in 34 discrete locations across 4 counties in southern England (Devon, Somerset, Norfolk and Suffolk) and 2 counties of north Wales (Anglesey, Caernarfonshire) in July 2006. Collection methods are detailed in full in Chapter 2 (see section 2.2). All larvae and pupae collected were link-reared through to the adult stage, whereupon adults were card-point mounted (see section 2.2.1.2). These pinned adult specimens and their associated larval and pupal exuviae serve as voucher specimens for this study and are held in the Natural History Museum, London. Adults of these link-reared specimens were identified to species using the morphological keys of Cranston et al. (1987) and Snow (1990) and used for subsequent molecular analysis.

77

4.3.2 Molecular identification

4.3.2.1 CQ11 Microsatellite Assay

The PCR for the amplification of the microsatellite locus CQ11 was carried out using a single leg from 322 specimens that were morphologically identified as Cx. pipiens s.l. The PCR was run in a 20µL reaction mix (Table 4.2) using the cycling conditions described by Bahnck & Fonseca (2006). PCR amplified fragments were visually differentiated on a 2% agarose gel, in comparison to a known size standard (HyperLadder IV, BioLine, UK) (Figure 4.2). After Bahnck & Fonseca (2006), a 180-bp fragment was scored as Culex pipiens f. pipiens, a 250-bp fragment was scored as Cx. pipiens f. molestus and the presence of both amplicons in a single sample was scored as a putative Cx. pipiens f. pipiens x f. molestus hybrid (Figure 4.2).

78

Reagents (concentration) Volume (20µl) Thermocycler conditions ddH20 12.4 1) 95°C- 10mins 10x NH4 buffer (BioLine) 2.0 10mg/ml BSA 0.3 2) 94°C- 30secs 10mM dNTPs 0.4 3) 55°C- 30secs 10µM P(CQ11mol) 0.2 4) 72°C- 40secs 10µM P(CQ11F) 0.2 Repeat steps 2-4 for 39 cycles 10µM P(CQ11pip) 0.2 25mM MgCl2 (BioLine) 1.6 5) 72°C- 5mins Taq (BioLine) 0.2 DNA 2.5

Table 4.2 Composition of the PCR reagents and thermocycler conditions used in the 20µl reaction for the CQ11 microsatellite assay of Bahnck & Fonseca (2006).

Reagents (concentration) Volume (25µl) Thermocycler conditions ddH2O 14.90 1) 95°C- 5mins 10x NH4 buffer (BioLine) 2.50 2) 95°C- 30secs 10mM dNTPs 0.50 3) 53.5°C- 45secs 10µM F primer (LCO) 1.25 4) 72°C- 45secs 10µM R primer (HCO) 1.25 Repeat steps 2-4 for 24 50mM MgCl2 (BioLine) 2.50 cycles Taq (BioLine) 0.10 5) 72°C- 4mins DNA 2.00

Table 4.3 PCR reaction mix and thermocycler conditions for amplification of the barcoding region of the COI gene (25µl reaction).

79

250bp 180bp

Figure 4.2 Image of the results of the CQ11 microsatellite assay following electrophoresis on a 2% agarose gel. Culex pipiens f. molestus (250bp) and Cx. pipiens f. pipiens (180bp) controls are clearly shown in Lanes 9 and 10 respectively. Lane 1: HyperLadder IV (BioLine), Lane 2: Culex pipiens f. pipiens, Lane 3: individual with fragments for both Cx. pipiens f. pipiens and Cx. pipiens f. molestus, Lanes 4-8: Culex pipiens f. molestus, Lane 9: Culex pipiens f. molestus control and Lane 10: Culex pipiens f. pipiens control.

80

4.3.2.2 Amplification of mtDNA COI gene fragment

Thirty samples previously identified in Section 4.3.2.1 as Culex pipiens f. pipiens (n=10), Cx. pipiens f. molestus (n=14) and pipiens x molestus hybrids (n=6) were analysed further using the mtDNA COI gene (Table 4.5). Instead of using the COI PCR-RFLP assay of Shaikevich (2007), the COI gene was sequenced and resultant fragments were screened for the Cx. pipiens f. molestus and Cx. pipiens f. pipiens diagnostic base change reported by Shaikevich (2007).

A 710-bp region of the mitochondrial Cytochrome Oxidase I (COI) gene (corresponding to the “barcoding” region) that overlapped the restriction sites of the COI RFLP assay (Shaikevich, 2007) was amplified using the primers LCO1490 and HCO2198 (Folmer et al., 1994). Volume and concentration of reagents used in the PCR reaction are given in Table 4.3. PCR products were cleaned in a 200-µl PCR tube using 8µl of positive PCR product and 2µl of a 1:4 dilution of ExoSAP-IT® (GE Healthcare, UK). The mixture was placed in a thermocycler and incubated at 37ºC for 30 minutes and then at 80ºC for a further 20 minutes. Cleaned products were then sent to the Zoological Sequencing Facility at the Natural History Museum for sequencing. All COI sequences were assembled and edited using Sequencher® 4.6 and aligned using ClustalW (Thompson et al., 1997). A maximum parsimony tree was constructed using PAUP 4.0 b10 (Swofford, 2002). Bootstrap values for 100 replicates were calculated using TNT (Goloboff et al., 2008) using a heuristic search of 100 replicates of TBR branch swapping. Genetic diversity (pi) within and p-distance between clades (see Results) was calculated using MEGA 4.0 (Tamura et al., 2007) and DnaSP 4.0.2 (Rozas et al., 2003).

As the COI-RFLP assay designed by Shaikevich was intended to distinguish the two forms of Cx. pipiens s.l. and Cx. torrentium, a further 18 COI sequences were obtained and analysed together with the 30 specimens listed: morphologically verified Cx. torrentium specimens (collected in this study, n=10), Cx. pipiens f. molestus colony individuals from Greece (n=3) (supplied by G. Koliopolous, Benaki Phytopathological Institute, Athens Greece; sequenced by Y. -M. Linton), 1 wild-caught Cx. pipiens f. molestus from Barking, London (supplied by A. Curtotti, Queen Mary University), GenBank sequences of Cx. pipiens f. pipiens (accession number AM403476, n=1), Cx. pipiens f. molestus (AM403492, n=1) and 81

Cx. torrentium (AM403477, n=1) from Russia (Shaikevich, 2007) and a sequence from colony Cx. quinquefasciatus of the Pipiens Group (n=1) (supplied by Prof. A.J. Mordue, University of Aberdeen, Scotland; sequenced by Y.-M. Linton). The closely related Cx. vagans (n=4) (sequenced by Y. -M. Linton) belonging to the Trifiliatus subgroup in the Pipiens Group and Cx. sitiens (n=1) from Australia (DQ673858) of the Sitiens Group were used as outgroup taxa.

4.4 Results

4.4.1 Collection of Culex mosquitoes

Aquatic stages of Culex mosquitoes collected only from overground breeding sites in this study (n=427) were morphologically identified as Cx. pipiens s.l. (n=322) or Cx. torrentium (n=105). Culex pipiens s.l. was detected in all counties sampled (Anglesey (n=105) and Caernarfonshire (n=21) in Wales, Devon (n=139), Norfolk (n=27), Somerset (n=24) and Suffolk (n=6) in England). Culex torrentium was not detected in Somerset or Suffolk, but was present in the other counties sampled [Anglesey (n=34), Caernarfonshire (n=14), Devon (n=45) and Norfolk (n=12)]. Neither of the other two reported British species of Culex (Cx. europaeus and Cx. modestus) was collected in this study.

4.4.2 CQ11 Microsatellite assay

The 322 Culex mosquitoes were analysed using the assay of Bahnck & Fonseca (2006) (Table 4.4). Culex pipiens f. molestus was detected in all areas sampled but generally in lower numbers than Cx. pipiens f. pipiens (Table 3.4). Both forms were detected sympatrically in 72.5% sites. Hybrids of pipiens x molestus comprised 6.8% (n=22) of all Cx. pipiens s.l. tested in this study. These twenty-two heterozygous individuals were found in 9 of the 40 sites. This level of hybridization and the detection of overground populations of Cx. pipiens f. molestus were highly surprising; therefore, verification of these results was sought using the PCR-RFLP method of Shaikevich (2007).

82

County Cx. pipiens f. molestus Cx. pipiens f. pipiens ‘hybrid’ Anglesey 49 49 7 Caernarfonshire 5 13 3 Devon 31 96 12 Norfolk 3 24 - Somerset 5 19 - Suffolk 2 4 - Total 95 205 22

Table 4.4 Identification of Cx pipiens s.l. specimens using the assay of Bahnck & Fonseca (2006); Cx. pipiens f. molestus (n=95), Cx. pipiens f. pipiens (n=205), hybrids (n=22).

COI (n=40) Morphology CQ11 (n=30) Cx. pipiens f. Cx. torrentium pipiens Cx. pipiens s.l. (n=30) 10 Cx. pipiens f. pipiens 7 3 14 Cx. pipiens f. molestus 7 7 6 pipiens x molestus hybrid 0 6 Cx. torrentium (n=10) 0 10

Table 4.5 A total of 30 specimens (from Table 3.4): Cx. pipiens f. pipiens (n=10), Cx. pipiens f. molestus (n=14) and pipiens x molestus hybrid (n=6) and ten morphologically identified Cx. torrentium were sequenced for COI. Upon analysis, the assay of Shaikevich (2007) distinguished 14 Cx. pipiens f. pipiens individuals, 26 Cx. torrentium individuals and no Cx. pipiens f. molestus or hybrid specimens were detected.

83

4.4.3 Congruence of CQ11 assay and mtDNA COI sequences

Three separate clades emerged in the tree reconstruction using maximum parsimony: pipiens/molestus, torrentium and Cx. vagans with Cx. sitiens as the outgroup (Figure 4.3). The pipiens/molestus clade included 14 specimens of purported Cx. pipiens f. pipiens (n=7) and Cx. pipiens f. molestus (n=7) samples that were identified using the CQ11 assay of Bahnck & Fonseca (2006). It also included the control specimens of Cx. pipiens f. pipiens and Cx. pipiens f. molestus from Russia (Shaikevich, 2007), Cx. pipiens f. molestus from London (n=1) and Cx. pipiens f. molestus colony specimens from Greece (n=3) (Figure 4.3). Only three unique haplotypes were recovered in the 21 COI sequences, each differing by a single base. Firstly, the sequence of Cx. quinquefasciatus from the University of Aberdeen colony showed 100% identity to the 3 Cx. pipiens f. molestus colony specimens from Greece, which could be due either to laboratory or colony contamination. The control sequence of the Russian Cx. pipiens f. molestus (AM403492; Shaikevich, 2007) was identical to the one wild- caught specimen from sewage tunnels in London. Finally, an identical COI haplotype was shared between the Russian Cx. pipiens f. pipiens (AM403476; Shaikevich, 2007) and UK samples identified as Cx. pipiens f. pipiens (n=7) and Cx. pipiens f. molestus (n=7).

The torrentium clade (n=27) (Figure 4.3) comprised the 10 morphologically verified samples as well as the Cx. torrentium of Shaikevich (2007) from Russia (AM403477). However, also in this clade were 16 specimens identified by the CQ11 assay as Cx. pipiens f. pipiens (n=3), Cx. pipiens f. molestus (n=7) and all 6 hybrids sequenced. Interestingly, a high level of intraspecific variation was noted in the torrentium clade (n=27; 13 haplotypes; Pi=0.0324), while diversity was extremely low in pipiens-molestus (n=21; 3 haplotypes; Pi=0.00095). This was despite the wider geographic origin of the specimens (Figure 4.3). Culex torrentium and Cx. pipiens s.l. COI sequences were 12.48% divergent on average, with 15 polymorphic sites, 6 of which appeared to be clade-specific. A comparison of COI sequences generated herein with those of Shaikevich (2007) showed that despite this intra- specific variation (9 variable sites), Cx. torrentium in the UK can be reliably identified using the markers included in the COI-RFLP assay of Shaikevich (2007); however, the purported unique restriction site for Cx. pipiens f. molestus is only present in one control specimen (from

84

London) and not in either Cx. pipiens f. molestus colony material or in the 7 other specimens identified as Cx. pipiens f. molestus using the CQ11 assay.

85

Figure 4.3 Maximum parsimony tree of the mtDNA Cytochrome Oxidase I gene (COI, 710bp) of Cx. pipiens s.l. and Cx. torrentium specimens (from Table 3.5). Bootstrap values are indicated above branches. Terminal labels show individual DNA numbers, species name, specimen origin or GenBank accession numbers for published sequences. Specimens labelled in orange were morphologically identified as Cx. pipiens s.l. and further distinguished as f. pipiens (n=10), f. molestus (n=14) and hybrids (n=6) using the CQ11 microsatellite assay of Bahnck & Fonseca (2006). Of these, 16 specimens occur within the torrentium clade together with morphologically identified Cx. torrentium (n=10, marked in Black), Cx. vagans and Cx. sitiens were included as outgroups (marked in blue), while Cx. pipiens f. molestus Greece, Cx. quinquefasciatus_Aberdeen both from colonies and f. molestus_london collected in sewage tunnels in Barking, London were used as positive controls (marked in blue).

86

4.5 Discussion

Species complexes are relatively common in the Family Culicidae, particularly in taxa that have large distributions and appear plastic in their ecological requirements (Harbach, 2004), e.g. Anopheles gambiae (White, 1985) and Ochlerotatus caspius (Schultz et al., 1986). The presence of cryptic taxa or the sympatric distribution of closely isomorphic species can make correct species identification problematic. In the UK, the presence of morphologically similar Cx. torrentium and Cx. pipiens s.l., combined with the fact that the latter is a species complex (Cx. pipiens f. pipiens and Cx. pipiens f. molestus), makes the identification process challenging. Two recently developed molecular assays differentiating Cx. pipiens f. pipiens and Cx. pipiens f. molestus were tested to determine their suitability and accuracy in detecting the presence of in the UK.

From this study, three interesting results were obtained. Firstly, according to the CQ11 assay of Bahnck & Fonseca (2006) Culex pipiens f. pipiens, Cx. pipiens f. molestus and hybrids of Cx. pipiens f. pipiens and Cx. pipiens f. molestus were detected overground in the UK samples. In fact, the presence of Cx. pipiens f. molestus was found repeatedly, in Devon, Somerset, Norfolk, Suffolk, Anglesey and northern Wales. This paradoxically constitutes the first record of overground presence of Cx. pipiens f. molestus in the UK and contradicts the eco-reports of it being found solely in underground habitats (e.g., Cranston et al., 1987). One reason for this apparent contradiction could be attributed to the assumption that the two forms are genetically distinct due to a lack of gene flow between overground and underground populations (Bryne & Nichols, 1999; Vinogradova et al., 2007); an assumption on which assays used to discriminate Cx. pipiens f. pipiens from Cx. pipiens f. molestus have been designed. However, the presence of overground Cx. pipiens f. molestus has also been recorded in sympatry with Cx. pipiens f. pipiens in Portugal (through the presence of autogeny only, Diaz et al., 2006) and in Russia (identification by COI only, Vinogradova et al., 2007). These findings imply that Cx. pipiens f. molestus is not restricted to underground habitats and that these ecological traits may not be as fixed as previously believed.

Secondly, the CQ11 assay identified 3 Cx. pipiens f. pipiens, 7 Cx. pipiens f. molestus and 6 hybrids that, based on the COI sequences obtained here, would have been identified as torrentium using the RFLP assay of Shaikevich (2007). This result strongly questions the use 87 of the absence of pre-alar scales as a morphological character for the identification of female Cx. pipiens s.l. from Cx. torrentium. The phylogenetic tree of COI contained individuals without pre-alar scales (Cx. pipiens s.l.) within the Cx torrentium clade, further indicating that absence of pre-alar scales is not a reliable diagnostic character. Given that 23% of the identified specimens collected in this study were females and that wild-caught specimens (even those link-reared from immatures as herein) either do not all possess these scales, or they can be easily lost (for example due to rubbing of adult specimens), thus indicating a need for a more robust morphological character that can accurately differentiate females of the two species. Furthermore, the morphological misidentification of Cx torrentium as Cx. pipiens s.l. caused some confusion in interpreting the results of CQ11 assay as one or both fragments were produced by specimens of this species. This resulted in either a molecular mis- identification of Cx. torrentium as Cx. pipiens f. molestus or the appearance of what appeared to be f. pipiens x f. molestus hybrids. Despite the fact that closely related mosquito species are known to hybridise (Spielman 1967; Takai & Kanda, 1986; Taylor, 1988; Urbanelli et al., 1997) there is no evidence for this in the UK. Taken together, these findings highlight serious shortfalls in the correct identification of Cx. torrentium and Cx. pipiens s.l. using either morphology or the CQ11 assay.

Finally, further analysis of the COI data showed that 7 individuals that were identified by the microsatellite assay of Bahnck & Fonseca (2006) as Cx. pipiens f. molestus were, in fact, Cx. pipiens f. pipiens according to the COI data. COI sequences obtained from individuals identified by the CQ11 assay were clustered into 2 clades: pipiens/molestus and torrentium. The single base (A-G, 205 bp) polymorphism was reported to be diagnostic between Cx. pipiens f. pipiens (A) and Cx. pipiens f. molestus (G) (Shaikevich, 2007). The G nucleotide (i.e., Cx. pipiens f. molestus) was observed in underground collections of Cx. pipiens s.l. in London as well as in the Russian specimens, both used here. However, this specific G nucleotide was not present in either Cx. pipiens f. molestus colony material from Aberdeen or from Greece (Figure 3.5). Thus according to the COI-RFLP of Shaikevich (2007), the colony material would have been misidentified as Cx. pipiens f. pipiens, implying that collections from London and from Russia are “true” Cx. pipiens f. molestus specimens and that the colony specimens from Greece and Aberdeen are autogenous Cx. pipiens f. pipiens. This however is contradictory, as autogeny is the single ecological character defining Cx. pipiens f. molestus. Interestingly, this result raises the possibility that these Cx. pipiens f.

88 molestus specimens are in fact hybrids with a Cx. pipiens f. pipiens mother (assuming mtDNA was transferred from the female). However, the CQ11 assay was capable of detecting hybrids of the two forms (Bahnck & Fonseca, 2006), visualised by the presence of two fragments (corresponding to each form) on an agarose gel as was seen in the putative hybrids that were, in fact, Cx. torrentium (see above). Thus on comparison of the results obtained from these two assays, it is apparent that neither the microsatellite (CQ11) nor the COI-RFLP assay is able to reliably differentiate the Culex pipiens complex in the UK.

One plausible reason for the inability for the COI to differentiate Cx. pipiens f. pipiens and Cx. pipiens f. molestus consistently could have arisen from the low genetic diversity observed within the pipiens-molestus clade. The extremely low COI variation in Cx. pipiens s.l., compared to that noted in Cx. torrentium, suggests that differences observed in population level surveys do not hold up across a wider geographic area. In fact, the low level of genetic diversity within the pipiens-molestus clade 0.002% (2 bp in 710) brings into question the taxonomic validity of two ecological forms within Cx. pipiens s.l. Studies on identification of species based on known COI sequences have reported low level variation of up to 0.25% within species. Interestingly, the presence of this substantial variation observed within and between (from 0.25% within to 8 % between, Hebert et al., 2004) species in the COI gene has resulted in it being used as a “barcode” which has been shown to be highly successful in differentiating mosquito species to date (Cywinska et al., 2006; Kumar et al., 2007). Variation in COI sequences support the differentiation of Cx. torrentium from Cx pipiens s.l., while the low variation observed within the pipiens-molestus clade indicates a shared mitochondrial DNA lineage, indicative of a single species.

As mentioned earlier, the single character consistently differentiating the two forms is autogeny (Bryne & Nichols, 1999; Bahnck & Fonseca, 2006; Diaz et al., 2006; A. Curtotti, pers. comm.). Given that both molecular assays identified a proportion of overground individuals as Cx. pipiens f. molestus, both molecular markers may be inefficient in the identification of Cx. pipiens s.l. Autogeny is a genotypic trait controlled by two chromosomes in mosquitoes (Spielman, 1957) and its inheritance is reported to be non-Mendelian in nature (Krishnamurthy, 1961). The expression of autogeny in other species of mosquitoes is reported to be flexible in relation to fluctuating environmental pressures. Examples include the presence of carbohydrates for adults (Su & Mulla, 1997) or increased larval nutrition (O’

89

Meara & Kranjsick, 1970) in Aedes atropalpus as well as Cx. quinquefasciatus (Olejnick & Gelbic, 2000). Thus expression of autogeny in Cx. pipiens s.l. as a result of epigenetic influences should also be considered, as it could imply variability within a single species.

Increased levels of methylated DNA present in Cx. pipiens f. molestus could explain expressed autogeny in mosquitoes that are found in small confined environments. Methylated DNA is present in 0-3% of an insect genome (Field et al., 2004). It is thought to influence the transcription and expression of specific genes that may otherwise be lost (Mandrioli, 2004). Field et al. (1989) showed elevated levels of methylated cytosine in two esterase genes present in insecticide resistant Myzus persicae (peach potato aphid) and Hick et al. (1996) found that the presence of methylated cytosine decreased with reduced expression of two esterase genes. These studies demonstrate the effect of environmental pressures on gene expression. With such varying levels of autogenic expression, it is proposed that the expression of autogeny could be induced in overground Cx. pipiens f. pipiens resulting in the seeming underground Cx. pipiens f. molestus.

In conclusion, there is no molecular evidence presented in this data set to suggest that the two forms of Cx. pipiens s.l. can be successfully differentiated based on either the CQ11 Bahnck & Fonseca (2006) assay or COI PCR-RFLP assay (Shaikevich, 2007). Furthermore, the lack of morphological characters, the incongruence of results from molecular assays, the lack of distinct overground and underground forms and the possibility of epigenetically induced expression of autogeny in Cx. pipiens f. molestus strongly suggest Cx. pipiens s.l. is one species. Therefore, all results in this study supports the suggestion of Harbach et al. (1984) that Cx. pipiens is a single, phenotypically plastic species that is easily adapted to a variety of habitats. Further studies on natural host selection and ecological parameters will be reported for both Cx. torrentium and Cx. pipiens s.l. in Chapters 4 and 5, respectively.

90

Chapter 5

Host selection in British mosquitoes

91

5. Host selection in British mosquitoes

5.1 Introduction

5.1.1 Mosquito-borne diseases in Europe

The emergence of mosquito-borne diseases such as Chikungunya (Rezza et al., 2007), the re-emergence of West Nile virus (WNv) into Europe (Hannoun et al., 1964; Tsai et al., 1998; Hubalek & Halouzka, 1999; Platonov et al., 2001; Del Giudice et al., 2004; Higgs et al., 2004; Esteves et al., 2005) and the predicted re-establishment of human malaria in the UK due to climate change (Snow, 1999; Medlock et al., 2005), have prompted entomologists to establish the potential of British species as vectors of mosquito-borne disease (Higgs et al., 2004; Medlock et al., 2005; Gould et al., 2006). To date, the isolation of WNv from An. maculipennis s.l. in Portugal (Filipe, 1972, in Esteves et al., 2005), Culex modestus in southern France (Hannoun et al., 1964), Culex pipiens s.l. in Romania (Tsai et al., 1998; Higgs et al., 2004) and Aedes vexans in Russia (Fyodorova et al., 2006) have incriminated these species as vectors. In addition to WNv, the isolation of other viruses occurring in northern and eastern Europe has implicated Aedes vexans, Cx. torrentium and Cx. pipiens in the transmission of Tahyna (Lundström et al., 2001), Sindbis (Jaenson et al., 1986b) and Usutu and Sindbis (Lundström, 1999) viruses, respectively. All of these species have been recorded in the UK (Snow, 1990) and evidence of local transmission of Sindbis and Usutu, between British mosquitoes and birds has been reported (Buckley et al., 2003)

5.1.2 Importance of host selection

Understanding mosquito host selection is essential for the accurate determination of potential bridge and secondary vectors (Lee et al., 2002) and for identifying vertebrate reservoir hosts (Lee et al., 2002; Oshagi et al., 2006). Host selection, as defined by Boreham & Garrett-Jones (1973), is the pattern of feeding observed through the analysis of specific blood meals in a mosquito population within a defined space and time. Host selection has been widely acknowledged as a means of understanding the relationships between hosts and vectors (Janini et al., 1995; Kilpatrick et al., 2007). Host preference, the preferred choice of a host as a

92 food source for haematophagous insects (Boreham & Garrett-Jones, 1973), can be studied directly by collection of the insects while feeding (Service, 1971) or by using host-specific odour-baited traps (Service, 1969). However, this preference may not be indicative of actual host selection because mosquitoes may feed opportunistically due to lack of available preferred hosts.

A host-seeking female often imbibes a single uninterrupted (unmixed) blood meal, but host irritability or an incomplete meal by the female may cause feeding to be interrupted (Davies, 1990). This can result in multiple meals, where two or more feeds are taken in a single gonotrophic cycle (Boreham & Garrett-Jones, 1973; Boreham, 1975; Romoser et al., 1989). Interruption of feeding increases host-vector contact thereby increasing the likelihood of pathogen transmission from the host to the mosquito and onto other hosts (Spielman, 1986; Beach et al., 1985). This capability of a female mosquito to transmit the pathogen amongst different host species increases its potential to be a bridge vector. Thus direct analysis of the mosquito blood meal has been considered the best way to directly assess host selection and multiple feeding under natural conditions (Boakye et al., 1999; Lee, et al., 2002).

5.1.3 Host identification

Host selection of mosquitoes has traditionally been determined through immunological tests, such as precipitin and Enzyme Linked Immunosorbent Assay (ELISA) tests. Both assays rely on the reaction between the host-specific serum from a blood meal and antibodies raised against that serum (Pant et al., 1987). The large amounts of antisera needed means these tests can be expensive and highly labour-intensive (Pant et al., 1987; Lee, et al., 2002). Aside from that, multiple assays are required to determine the specific host source (Burkot & DeFoliart, 1982) which can also compromise the specificity and sensitivity of both assays when heterologous sera or mixed meals are tested (Pant et al., 1987; Beier et al., 1988).

Molecular analysis of vector blood meals was first carried out on the tick, Ixodes ricinus, a major European vector of Lyme disease (Kirstein & Gray, 1996). Utilising a Restriction Fragment Length Polymorphism (RFLP) assay, exploiting fixed inter-specific mutations on the rapidly evolving mitochondrial Cytochrome B (CytB) gene, the authors were able to successfully identify the following eleven hosts: common mouse, wood mouse, bank

93 vole, sheep, red deer, silka deer, cow, rabbit, dog, fox and pheasant. Since then, the CytB gene has been the focus of several assays used to accurately discern vertebrate hosts in insect blood meals (Boakye et al., 1999; Kent & Norris, 2000; Meece et al., 2005; Oshagi et al., 2006). The high representation of vertebrate CytB sequences in Genbank further warrants the use of the CytB, as the sequences can be directly compared to the database for identification of the hosts to the species level.

5.1.4 British mosquitoes

Host selection of nearly all species recorded in the UK is known (Table 5.1). However in relation to vector potential, only the Anopheline mosquitoes are well studied. The transmission of human malaria in the British Isles has been recorded since the early 14th Century as ‘ague’, ‘tertiary’ or ‘quaternary’ fever (Reiter, 2000). Species incriminated include Anopheles atroparvus (Curtis & White, 1984) and An. plumbeus (Shute, 1954; Curtis & White, 1984; Marchant et al., 1998; Eling et al., 2003). The respective roles of An. messeae and An. daciae are uncertain, but interior malaria transmission in un-forested areas was historically attributed to An. messeae (Snow, 1990) and the species has been incriminated in other parts of its range (Detinova, 1953; Bruce-Chwatt & de Zulueta, 1980; Nikolaeva, 1996).

The most comprehensive studies on natural feeding behaviour of non-Anopheline British mosquitoes were carried out on mosquitoes collected in Poole, Dorset (Service, 1968b; 1971) (Table 5.1). Host selection was determined based on direct collection of insects from various hosts and immunological assay of the imbibed blood meal using a precipitin test (Table 5.1). Of the sixteen species whose blood meals were precipitin-tested by Service (1971), five (Ae. cinereus, Oc. caspius, Oc. punctor, Cx. torrentium and Cs. annulata) were shown to have fed on multiple vertebrate hosts in a single gonotrophic cycle (Table 5.1). In addition, Ramsdale & Snow (1995) reported that 25 of the 33 species of British mosquitoes had been recorded biting man (Table 5.1). Based on this and the status of these species as vectors in Europe and North America, Medlock et al. (2005) suggested that principal bridge vectors of WNv in the UK could include Cx. pipiens f. pipiens, Cx. torrentium, Cs. litorea and Cs. morsitans, while Ae. cinereus, An. plumbeus, Cs. annulata, Cs. litorea, Cs. morsitans, Cq. richiardii, Oc. cantans, Oc. detritus, Oc. dorsalis and Oc. punctor could also play a role as potential bridge vectors, but to a lesser degree. In a separate study, the isolation of WNv

94 antibodies from 46-day old chicks Buckley et al. (2006), suggests that British mosquitoes are involved in the local transmission of WNv in bird populations in the UK, yet no species can be definitely incriminated. In addition to potential WNv transmission, avian malaria was reported in Bristol Zoo (resulting in the death of 8 penguins), Edinburgh Zoo and in Marwell Zoo, (where 27 penguins succumbed to infection) (BBC news online, 1999). As penguins are non- endemic to the UK, transmission of avian malaria must occur from native birds to penguins

95

Mosquito species Preferred host Multiple meals (Direct feeding) (Precipitin test) An. algeriensis Man2,3,4 An. atroparvus Man2,4, Cattle3, Rabbit6 An. claviger Cattle1, Rabbit1, Mammals2, Man4 An. daciae ? An. messeae Man2,4 An. plumbeus Man2,3,4,5, Cattle1, Bird1, Mammals5 Ae. cinereus Cattle1,5, Bird1,5, Mammals2, Man4,5 Cow & Human1 Ae. geminus ? Ae. vexans Man2,4 Cs. alaskaensis Man2 Cs. annulata Bird1,2,5, Mammals2, Man2,4,5 Bird & Rabbit1 Cs. fumipennis ? Cs. litorea Bird1,2,5, Cattle1, Mammal2, Man2,4 Cs. longiareolata Bird1,2 Cs. morsitans Bird1,2,5, Man4,5 Cs. subochrea Man4 Cq. richiardii Bird1,5,9, Cattle1,5, Mammals2, Man4,5, Amphibians8 Cx. europaeus Amphibian/Reptiles2, Man4 Cx. modestus Man2 Cx. pipiens f. pipiens Bird1,2,5 f. molestus Man2,4,5, Birds5 Cx. torrentium Bird2,5 Bird & Rabbit1 Da. geniculata Man2,4 Oc. annulipes Mammals2,3, Man4 Oc. cantans Cattle1,5, Bird1,5, Man3,4,5, Mammals2,3,6 Oc. caspius Man2,4, Cattle1, Sheep1, Bird1 Bird & Mammal1 Oc. communis Man2,4 Oc. detritus Cattle1, Bird1, Man2,4 Oc. dorsalis Cattle1,5, Rabbit1, Pig1, Horse1, Mammals2,3, Man3,4 Oc. flavescens Cattle1,3, Sheep1,3, Bird1, Mammals2,3, Man4, Horses3 Oc. leucomelas Man2 Oc. punctor Man2,3,4,5, Cattle1,5, Bird1,5, Bird & Mammal1 Oc. rusticus Man2,3,4 Oc. sticticus Man2, Mammals7 Or. pulcripalpis Bird2

Table 5.1 Summary of available data on known host selection and multiple feeds in British mosquitoes (1Service, 1971; 2Snow, 1990; 3Cranston et al., 1987; 4Ramsdale & Snow, 1995; 5Medlock et al., 2005; 6Muirhead- Thompson, 1956; 7Mattingly, 1950; 8Shute, 1933; 9Service, 1969).

96

via endemic mosquitoes. It is uncertain, however, as to which species of endemic mosquitoes are involved in the transmission of avian malaria in the UK.

Given the presence of historical vectors of malaria in the UK and the role that local mosquitoes could play as vectors of emerging diseases such as WNv and malaria makes the introduction and establishment of these diseases a high possibility (Higgs et al., 2004, Medlock et al., 2005; Gould et al., 2006). Thus the identification of potential vectors and bridge vectors based on host identification and presence of parasitic infection in field-caught bloodfed females could elucidate the understanding of host-vector interactions in the UK and thereby elucidating the importance that these mosquitoes could have on human and animal health in the future.

5.2 Aims

The aims of this study were:

[1] To identify specific vertebrate hosts of British mosquitoes through the molecular analysis of blood meals in wild-caught mosquitoes collected from five counties in the UK,

[2] To identify candidate species that could act as vectors and bridge vectors of both human and animal diseases in the UK.

5.3 Materials and Methods

5.3.1 Collection and identification of blood-fed females

Resting blood-fed females were collected manually from the walls of animal shelters, derelict buildings, sheds and pillboxes in the English counties of Devon, Somerset and Norfolk (England) and in Anglesey (Wales) in July 2006 and in Kent (England) in August 2006 (Table 5.2) At two of the sampling sites, bloodfed females were also captured using Mosquito Magnet® traps (Liberty Pro) (Table 5.2; Bird World, Anglesey and Pettits Animal Farm, Norfolk). Once collected, blood-fed mosquitoes were visually scored according to their

97 degree of blood meal digestion as detailed in Figure 5.1(a-d). Gravid and non-bloodfed females were also recorded. Adults were carefully labelled with unique collection numbers and either pinned on card points, or placed in individual Beem® capsules and dried over silica gel (see Chapter 2, section 2.2.1.2 and 2.2.2). Mosquitoes were mostly identified to species using the British identification keys of Cranston et al. (1987) and Snow (1990). Members of the Maculipennis Complex were identified to species using the ITS2 PCR-RFLP assay designed in Chapter 3 and Cx. pipiens were treated as a single species (Chapter 4).

98

a b c d

Figure 5.1 Schematic diagram of a female mosquito showing the abdomen (a) fully (3/3) blood fed, (b) 2/3 bloodfed, (c) 1/3 bloodfed and (d) non–bloodfed or gravid. Mosquito outline from http://www.pestworldforkids.org/mosquitoes.html

99

Exact locality Date Collection Adult habitat Exact site Co-ordinates method (Deg. Decimal) Wales: Anglesey, Bird World (nr 08-09.07.06 Trap Inside shed near aviaries 53.1638428'N, 4.3316828'W Newborough) England: Devon, Exminster 06.07.06 Resting Brick Shelter 1 50.6752617'N, 3.4751598'W Marshes Nature Reserve 06.07.06 Resting Brick Shelter 2 50.6756321'N, 3.4743221'W

England: Kent, Cliffe Marshes 11.08.06 Resting Brick shelter 51.4671890'N, 0.4895240'E 11.08.06 Resting Sheep Corral 51.4783052'N, 0.4827834'E England: Norfolk, Pettits Animal 16.07.06 Resting Goat stables 52.5661404'N, 1.5768701'E Farm* (nr Reedham) 16.07.06 Resting Reindeer stables 52.5659609'N, 1.5768554'E 16.07.06 Resting Mini horse/donkey stables 52.5659609'N, 1.5768554'E 16-17.07.06 Trap Near pond in petting area 52.5661226'N, 1.5774591'E 17-20.07.06 Trap Near bird aviaries 52.4656741'N, 1.5704780'E 20.07.06 Resting Goat stables 52.5661404'N, 1.5768701'E 20.07.06 Resting Rhea stables 52.5659609'N, 1.5768554'E 20.07.06 Resting Reindeer stables 52.5659609'N, 1.5768554'E 20.07.06 Resting Mini horse/donkey stables 52.5659609'N, 1.5768554'E England: Somerset, Godney 07.07.06 Resting Pillbox 1 51.1816487'N, 2.7273050'W Farm, Godney 07.07.06 Resting Horse stables 51.1817652'N, 2.7230142'W 07.07.06 Resting Pillbox 2 51.1822719'N, 2.7283164'W

Table 5.2 List of habitats, collection dates and co-ordinates (in degree decimal) of sites where all adult blood fed and resting mosquitoes were collected for this study. Resting adults were collected in Devon, Kent, Norfolk and Somerset, while host-seeking adults were collected in Anglesey and Norfolk. *Collections were taken from the low open stables of reindeer and miniature donkeys and ponies in adjacent pens of the Petting Area of Pettits Animal Farm.

.

100

5.3.2 DNA extraction protocols

5.3.2.1 DNA extraction from host serum and dried blood

DNA extracts from serum samples (Sera Laboratories International, West Sussex, England) from cow, dog and horse were used as controls for the optimisation of universal CytB primers and for testing the specificity of designed primers (herein). Proteinase K (20µl) and 200µl of buffer AL (QIAgen®) was added to 200µl of the control animal serum in a 1.5- ml Eppendorf® tube. The sample was mixed by vortexing for 15 seconds and placed in a heat block at 56ºC for 15 minutes. The sample was briefly centrifuged and 200µl of 100% ethanol was added and mixed by vortexing. The mixture was then transferred to a QIAgen® spin column and DNA was extracted using the QIAmp Mini Blood Kit following the manufacturer’s instructions for extraction from human blood. Control human DNA was extracted from dried blood smears on filter paper; a 1-cm2 piece was cut from the filter paper and placed in a 1.5-ml Eppendorf® tube to which 180µl of ATL (QIAgen®) was added. The tube was placed in a rotator and left overnight. DNA was then extracted according to the protocol described above.

5.3.2.2 DNA extractions for mosquito blood meal analysis

Abdomens of engorged females were separated from the head and thorax using a clean pair of forceps and a scalpel and placed individually in a clean 1.5ml Eppendorf® tube. The abdomen was then ground using a battery-operated pestle in 100ml of grinding buffer and DNA extracted according to the phenol-chloroform protocol of Linton et al. (2001b). Extracted genomic DNA (from both the mosquito and blood meal) was resuspended in a final volume of 100µl and stored at -20ºC, prior to PCR amplification.

101

5.3.3 PCR amplification protocols

5.3.3.1 Universal Cytochrome B Oxidase (CytB) amplification

The PCR protocol of Boakye et al. (1999) (Table 5.4), employing their universal CytB primers (CytB-F and CytB-R), was optimised using 21 field-caught female mosquitoes as follows: fully bloodfed (n=13), 2/3 bloodfed (n=2), 1/3 bloodfed (n=2), gravid (n=2) and unfed (n=2) from Norfolk and Somerset. PCR fragments of 350bp were obtained, using the PCR conditions listed in Table 5.4. PCR products were cleaned in a 200µl PCR tube using 8µl of positive PCR product and 2µl of a 1:4 dilution of ExoSAP-IT® (GE Healthcare, UK). The mixture was placed in a thermocycler and incubated at 37ºC for 30 minutes and then at 80ºC for a further 20 minutes. Cleaned products were then sent to the Zoological Sequencing Facility in the Natural History Museum for sequencing. Resultant sequences were assembled and edited in Sequencher® version 4.6 (Gene Codes Corporation), aligned with ClustalW (Thompson et al., 1997) and compared to those available in GenBank using BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

102

DNA Optimised primer fragment concentrations and Primer Primer sequence (5'-3') (bp) annealing temperatures Cow-R GGAATGGGATTTTGTCTACATATGAGG 620 0.3µM, 48ºC Deer-R GGTGTATGATCCGTAGTATAGGCC 250 0.3µM, 51ºC Dog-R CAGTTCCGATATAAGGGATGGCAGAG 450 0.5µM, 51ºC Goat-R TATGAATGCTGTGGCCATTGTCGCGAGC 350 0.3µM, 51ºC Horse-R GGAGAGGATTAGGGCTAATACGCCG 800 0.3µM, 48ºC Human-R TCGAGTGATGTGGGCGATTGA 200 0.3µM, 51ºC CytB-R CCCTCAGAATGATATTTGTCCTCA 350 0.2µM, 60oC

Bird-F TACAAAAAAATAGGCCCCGAAGG 650 0.2µM, 51ºC

Table 5.3 Host-specific primers designed in this study to complement the CytB-F primer (5΄- CCATCCAACATCTCAGCATGATGAA-3΄ of Boakye et al. (1999), showing expected fragment sizes and optimal primer concentrations and annealing temperatures. Bird-F (designed by Y.-M. Linton) was designed to pair with the CytB-R primer of Boakye et al. (1999).

Reagents Volumes Thermocycler conditions (25µl) reaction ddH2O 18.15 1) 95°C for 3.5mins 10x NH4 buffer 2.50 2) 95°Cfor 50 secs 10mM dNTPs 0.50 3) 51°C for 50secs 10µM CytB-F 0.50 4) 72°C for 40secs 10µM host-R 0.50 Steps 2-4 repeated for 34 cycles 50mM MgCl2 0.75 5) 72°C for 5mins Taq 0.10 DNA 2.00

Table 5.4 Optimised PCR master mix and thermocycling conditions for the amplification of vertebrate CytB gene, using host-specific primers designed in this study [after Boakye et al. (1999)]. Reagents including 10x NH4 buffer, 50mM MgCl2 and Taq were from BioLine®.

103

5.3.3.2 Host-specific primer design and PCR optimisation

Seven potential vertebrate host groups were determined, based on their prevalence near collections sites in summer 2006, as follows: birds, cattle, deer (including reindeer), dogs, goats, horses and man. Where multiple sequences were available from GenBank, pairwise intra-group p-distances were calculated in PAUP* 4.0 b10 (Swofford, 2002) and the two most divergent sequences were then manually re-aligned using Sequencher® version 4.6. Sequences used to develop the host-specific forward primers (Table 5.3): Bos taurus (Cow; AB074963, AF490529), Capreolus capreolus (European Roe Deer: Y14951), Cervus elaphus (Red Deer: AJ000022), Dama dama (Red Water Deer: X56290I); Muntiacus reevesi (Muntjac Deer: EF035447), Rangifer tarandus (Reindeer: NC007703); Canis familiaris (Dog: NG002008); Canis lupus (Wolf: DQ480500); Capra hicrus (Goat: EU130780, AF217254); Equus caballus (Horse: NC001640, EF597512) and Homo sapiens (Man: NC001807, EU935442).

All species-specific reverse primers designed herein were developed to be paired with the universal forward primer of Boakye et al. (1999), producing amplified fragments of sufficiently differing sizes that each host species could be visually determined directly from an agarose gel. Expected product sizes are shown in Table 5.3. For birds, a bird-specific primer obtained from Y.-M. Linton (pers. comm.) was designed to work with the universal reverse primer of Boakye et al. (1999) and yielded a 600bp fragment, including a short fragment (50bp) of the ND5 region.

Primers were checked for complementing annealing temperature and the capability to form primer-dimers using an online oligo-analyser (http://eu.idtdna.com/analyzer/applications/oligoanalyzer/default.aspx). Optimal annealing temperatures (48-52ºC) and primer concentrations (1-5µM) for each primer pair (Table 5.3) were determined using the control host DNA as well as the relevant the universal CytB primer (section 5.3.3.1), in a temperature and primer concentration gradient PCR reaction. Controls for cow, dog and horse were obtained from serum, human controls from blood smears and controls for bird, deer and goat were obtained from blood meals sequenced for the optimisation of Universal CytB primers (See section 5.4.1).

104

Specificity of host primers was also tested against both target and unintended hosts using controls listed above. The production of bands for the target host and the absence of bands in all other host controls indicated primer specificity. To ensure accurate amplification of target host, 12 positive PCR products from control samples [bird (n=1), deer (n=1), dog (n=2), goat (n=3), horse (n=2) & man (n=3)] were cleaned and directly sequenced (see section 4.3.3.1) in both directions. Further to that, 20 positive PCR products from field caught bloodfed mosquitoes [bird (n=9), deer (n=1), dog (n=3) and goat (n=7)] were cleaned and directly sequenced (see section 4.3.3.1) in both directions, to ascertain accurate identification of hosts on bloodfed specimens using host specific primers. Resultant sequences were compared to those in GenBank using BLAST to confirm the identity of the host.

5.3.3.3 Determination of host selection

Single PCR reactions (see section 5.4.2) for all but the cow-specific primers (due to the lack of specificity of this primer), were carried out for each of the above primer pairs on the DNA extracted from 280 mosquito abdomens to determine natural host selection. PCR was performed under conditions listed in Table 5.4. Fragments were visualised following electrophoresis of 4µl of PCR reaction on a 1.5% agarose gel at 70V containing 1% ethidium bromide. Fragment sizes were measured against Hyperladder IV standards (BioLine®). In cases where the host-specific primers did not yield any PCR product, the DNA was re- amplified using the universal CytB primers of Boakye et al., (1999) (Table 5.3) and subsequently sequenced.

In order to ensure the accuracy of the host-specific primers when applied to blood meals from field-caught adults, PCR products were sequenced from 20 bloodfed mosquitoes that had been shown to feed on birds (n=9), deer (n=1), dogs (n=3) and goats (n=7) (for protocol see section 5.3.3.1). Results were compared with reference sequences in GenBank using BLAST. Sequences were edited and aligned as above and a phylogenetic tree was constructed using a maximum likelihood search under a GTR model of nucleotide substitution with empirical base frequencies. Node support was calculated with nonparametric bootstrap (100 reps). Analyses were performed using PhyMLv2.4.4 (Guindon & Gascuel, 2003).

105

5.4 Results

Blood-fed female mosquito specimens from seven species were collected mainly as resting adults (n=668) and incidentally as host seeking females (n=119). Of the resting adults collected hosts from 280 specimens were identified (Table 5.5).

5.4.1 Optimisation of Universal CytB primers

DNA sequences of PCR products obtained in the optimisation of the universal CytB primers were recovered from all 21 field-caught mosquitoes [bloodfed (17), unfed (2) & gravid (2)]. Following blast searches with the resultant DNA sequences, four host species were detected in 12 of the 17 bloodfed individuals (70.6%), as follows: Capra hircus (goat: An. maculipennis s.l., n=2), Homo sapiens (man: Cq. richiardii, n=1), Rangifer tarandus (reindeer: An. daciae, n=5; An. messeae, n=3) and Pterocnemia pennata (Darwin’s Rhea: An. daciae, n=1). The 9 remaining DNA sequences produced from gravid (n=2), unfed (n=2) and bloodfed (n=5) mosquitoes showed 99-100% similarity to Armigeres subalbatus, indicating that the universal primers also amplified mosquito DNA. Subsequently vertebrate host- specific primers were designed and used in widespread screening for host selection in this study.

106 County (n=280) Adult habitat Species Bloodfed females Proportion bloodfed (n=280) Anglesey (n=1) Shed Cx. pipiens 2/3 (1) Devon (n=58) Brick Shelter 1 An. messeae 1/3 (7), 2/3 (4) Brick Shelter 2 An. messeae 1/3 (25), 2/3 (19), 3/3 (2) Maculipennis Gp 1/3 (1) Kent (n=36) Brick Shelter An. atroparvus 3/3 (2) Maculipennis Gp 2/3 (1) Derelict building used An. atroparvus 1/3 (25) as sheep corral An. daciae 1/3 (6) Maculipennis Gp 1/3 (2) Norfolk (n=148) Goat stables An. atroparvus 1/3 (1), 2/3 (5), 3/3 (1) An. daciae 1/3 (5), 2/3 (36), 3/3 (12) Maculipennis Gp 1/3 (3), 2/3 (10), 3/3 (4) An. messeae 2/3 (1) Cs. spp 3/3 (1) Near pond in Petting Cq. richiardii 3/3 (1) area Rhea stables An. daciae 3/3 (2) An. messeae 3/3 (1) Reindeer stables An. atroparvus 2/3 (1) An. daciae 2/3 (23), 3/3 (1) Maculipennis Gp 2/3 (2) An. messeae 2/3 (1) An. spp 2/3 (3) Mini horse/donkey An. daciae 2/3 (1), 3/3 (27) stables Maculipennis Gp 2/3 (1), 3/3 (2) An. messeae 3/3 (3) Somerset (n=37) Pillbox 1 An. daciae 1/3 (4), Gravid (2), Non-BF (1) Maculipennis Gp 2/3 (1) An. messeae 1/3 (1) Cx. pipiens 1/3 (1), 3/3 (2) Cx. torrentium 1/3 (1) Cs. subochrea 2/3 (2) Horse stables An. daciae 1/3 (7), 3/3 (1) Maculipennis Gp 1/3 (2) An. messeae 1/3 (1) Cq. richiardii Non-BF (1) Pillbox 2 An. daciae 1/3 (6), 3/3 (3) Cs. spp Non-BF(1)

Table 5.5 List of 280 bloodfed (BF) specimens used to analyse host selection and determine natural parasitic infection. Abdomens of bloodfed mosquitoes were scored visually (Figure 4.1a-d) as follows: 1/3 bloodfed, 2/3 bloodfed, 3/3 bloodfed, gravid and non-bloodfed (non-BF). Gravid and non-bloodfed specimens were included as negative controls.

107

5.4.2 Optimisation of host specific primers

Of the 7 host-specific primers, all but the cow primer, were shown to amplify the target host CytB gene when tested on known controls: bird (n=1), human (n=3), horse (n=2), goat (n=3, dog (n=2) & reindeer (n=1). Due to the lack of specificity, the designed cow primer was not used in subsequent analysis. No cross-contamination was otherwise noted. Following optimisation, host-specific primers were then used for screening the 280 wild- caught specimens.

Out of the 280 field caught bloodfed mosquitoes analysed, twenty specimens [bird (n=9), deer (n=1), dog (n=3) and goat (n=7)] were randomly sequenced to ascertain the accurate identification of hosts using the designed host specific primers.

Of these, nine specimens positive with the bird-specific primer, four (all An. daciae) had 100% sequence similarity to P. pennata [Figure 4.2 (Darwin’s rhea) from Pettits Animal Farm, Norfolk] and one Cx. torrentium from Somerset shared 100% similarity to Fringilla coelebs (Figure 5.2b) From the remaining four specimens (An. messeae, n=1 and An. daciae, n=3) sequence data from a 50bp mitochondrial ND5 fragment was recovered that was similar to that of Lanius meridionalis (grey shrike) (AM494443). Interestingly, this species of bird is neither endemic to the UK nor present in the environs of the Pettits Animal Farm in Norfolk and was only detected in these individuals. The blood meal of one An. daciae collected at the same site showed 95% similarity to C. elaphus (Red Deer, Figure 5.2b).

Of the seven DNA sequences generated from blood meals amplified using the goat- specific primer, six of showed 100% similarity to C. hircus (Figure 5.2b) (all An. daciae). However, the remaining single sequence generated from An. daciae with the goat-specific primer and the blood meal of all three Cx. pipiens specimens amplified using the dog primers shared 100% similarity to the mosquito Armigeres subalbatus (Figure 5.2a), despite producing good clean sequences. This suggests that in some cases, the dog primer and the goat primer also amplifying mosquito DNA.

108

Figure 5.2a

109

Figure 5.2b

Figure 5.2 A Maximum likelihood tree, with bootstrap values of nodes, of mtDNA Cytochrome B oxidase gene sequences. Terminal labels show DNA Number_species_Common name_degree Bloodfed_location or_Genbank accession numbers. Altogether there are 8 distinct clades: Dog (n=3, Genbank sequences=2), Deer (n=22, Genbank sequences = 6), Goat (n=10, Genbank sequences = 2), Cow (n=2, Genbank sequences= 2), Horse (n=1, Genbank sequences =2), Man (n=1, Genbank sequences=1), Bird (n=6, Genbank sequences =2), Mosquito (n=70, Genbank sequences =1).

Figure 5.2a shows bloodfed mosquito specimens from which no vertebrate hosts could be identified using host- specific primers. All specimens shown here were sequenced using Universal CytB primers and the closest match to the sequences obtained was Armigeres subalbatus (mosquito).

Figure 5.2b shows the hosts identified from bloodfed mosquito specimens. Coloured labels indicate samples sequenced with host specific primer: Orange-Deer (n=1), Green-Goat (n=3), Blue- Bird (n=5), brown- Dog (n=3), Black labels indicate samples sequences with Universal CytB primers (Boakye et al., 1999) when host specific primers did not work.

110

5.4.3 Host Selection

Of the 280 mosquitoes tested (including 5 negative controls), host selection was successfully determined for 166 (60.4%) specimens comprising seven UK mosquito species (Table 5.6). Blood meals were successfully amplified from mosquitoes 1/3 engorged through to fully engorged, of which 20 were sequenced to ascertain the accurate identification of hosts (Figure 5.2b). Hosts were identified from 134 specimens using the host-specific primers for bird, dog, deer, goat horse and man. Of the remaining specimens for which host-specific primers did not produce a result, universal CytB primers were used and sequences obtained. From this, host CytB sequences were generated from 32 individuals (Figure 5.2b) were identified while the remaining 109 specimens (39%) amplified mosquito DNA only. Interestingly, 4% of the all DNA fragments sequenced using the universal primer produced were ambiguous, showing double-peaks that were due either to the presence of amplified DNA from the mosquito itself, or possibly from an additional host. The majority of the specimens (n=141) were found to have had blood meals from a single host, whereas mixed blood meals were detected in 25 specimens (15%) (Table 5.6).

111

Locality Species sampled Unmixed meals (n=141) Mixed meals (n=25) (n=166) Bird Man Deer Goat Horse Cow Dog (17) (18) (43) (58) (2) (1) (2) Bird Cx. pipiens (1) 1 0 0 0 0 0 0 World, Anglesey Exminster An. messeae (40) 7 0 1 28 0 0 0 Bird/Goat (4) Marshes, An. maculipennis s.l. (1) 0 0 0 1 0 0 0 Devon Cliffe An. atroparvus (5) 0 0 4 1 0 0 0 Marshes, An. daciae (4) 0 0 3 1 0 0 0 Kent An. maculipennis s.l (3) 0 0 0 1 0 0 0 Man/Deer (1), Bird/Deer (1) Pettits An. atroparvus (5) 0 1 0 3 0 0 0 Man/Goat (1) Animal An. daciae (69) 4 11 29 14 2 1 0 Deer/Goat (2), Man/Goat (3), Man/Deer (2), Farm Bird/Man/Dog (1) Norfolk An. messeae (6) 0 1 3 0 0 0 0 An. maculipennis s.l. (17) 2 4 2 5 0 0 1 Bird/Goat (2) Bird/Deer (1), Man/Goat (1), Man/Deer (1) Cq. richiardii (1) 0 1 0 0 0 0 0 Godney, An. daciae (9) 3 0 0 4 0 0 0 Man/Bird (1), Bird/Deer (1) Somerset Cx. torrentium (1) 0 0 0 0 0 0 0 Bird/Dog (1) Cx. pipiens (3) 0 0 0 0 0 0 1 Man/Dog (1), Deer/Dog/Goat(1) Cs. subochrea (1) 0 0 1 0 0 0 0

Table 5.6 Host selection of British mosquitoes (n=166) successfully analysed in this study using species diagnostic primers designed herein and CytB DNA sequences.

112

Mosquito species Preferred host Multiple meals (Direct feeding) (CytB PCR assay) An. algeriensis Man2,3,4 An. atroparvus Man2,4, Cattle3, Rabbit6, Deer An. claviger Cattle1, Rabbit1, Mammals2, Man4 An. Daciae Man, Bird, Deer, Goat, Horse, Dog Bird & Man, Man & Goat, Man & Deer An. messeae Man2,4, Bird, Deer, Goat Bird & Goat An. plumbeus Man2,3,4,5, Cattle1, Bird1, Mammals5 Ae. cinereus Cattle1,5, Bird1,5, Mammals2, Man4,5 Cow & Human1 Ae. geminus ? Ae. vexans Man2,4 Cs. alaskaensis Man2 Cs. annulata Bird1,2,5, Mammals2, Man2,4,5 Bird & Rabbit1 Cs. fumipennis ? Cs. litorea Bird1,2,5, Cattle1, Mammal2, Man2,4 Cs. longiareolata Bird1,2 Cs. morsitans Bird1,2,5, Man4,5 Cs. subochrea Man4, Deer Cq. richiardii Bird1,5,9, Cattle1,5, Mammals2, Man4,5, Amphibians8, Cx. europaeus Amphibian/Reptiles2, Man4 Cx. modestus Man2 Cx. pipiens f. pipiens Bird1,2,5, Man, Dog Man & Dog Cx. torrentium Bird2,5 Bird & Rabbit1 Da. geniculata Man2,4 Oc. annulipes Mammals2,3, Man4 Oc. cantans Cattle1,5, Bird1,5, Man3,4,5, Mammals2,3,6 Oc. caspius Man2,4, Cattle1, Sheep1, Bird1 Bird & Mammal1 Oc. communis Man2,4 Oc. detritus Cattle1, Bird1, Man2,4 Oc. dorsalis Cattle1,5, Rabbit1, Pig1, Horse1, Mammals2,3, Man3,4 Oc. flavescens Cattle1,3, Sheep1,3, Bird1, Mammals2,3, Man4, Horses3 Oc. leucomelas Man2 Oc. punctor Man2,3,4,5, Cattle1,5, Bird1,5, Bird & Mammal1 Oc. rusticus Man2,3,4 Oc. sticticus Man2, Mammals7 Or. pulcripalpis Bird2

Table 5.7 Summary of available data as shown on page 109, Bold text shows the additional hosts detected in this study section 5.4.3.

113

5.5 Discussion

The study of host selection was carried out in this study as a means towards understanding the dynamics of disease transmission with specific reference to the recent resurgence and introduction of diseases such as WNv and Chikungunya beyond their range. Potential vector and bridge vector species in the UK have been identified based on previously published data for Europe and North America (Higgs et al., 2004; Medlock et al., 2005; see also Table 5.1) however, host selection and the vector potential of mosquito species in the UK have not been specifically tested in relation to these diseases. Herein host selection was determined, through the analysis of the mitochondrial CytB gene in blood meals of seven species of British mosquitoes (see Table 5.7) collected in (Devon, Norfolk, Somerset and the Isle of Anglesey): An. atroparvus, An. daciae, An. messeae, Cx. pipiens, Cx. torrentium, Cs. subochrea and Cq. richiardii. Seven host groups (bird, cow, dog, deer, goat, horse and man) were amplified from bloodfed female mosquitoes using host-specific primers designed herein for deer, dog, goat, horse and man as well as Universal CytB primers (Boakye et al., 1999).

The host selection of An. daciae is reported here for the first time. Both host-specific primers and CytB sequencing showed the host choices of An. daciae to be very indiscriminate. Six vertebrate hosts (bird, man, cow, deer, goat and horse) were determined from 84 individuals of An. daciae collected across 3 counties. That An. daciae is opportunistic in its feeding behaviour is exemplified in Norfolk where 71 bloodfed females, collected within the environs of the petting stables in Pettits Animal Farm, were shown to have fed on all six hosts present. Interestingly, An.messeae and An. atroparvus were far less diverse in their choice of hosts, despite being closely related to An. daciae. The selection of hosts for both An. messeae and An. atroparvus in Pettits Animal farm, where all three species were collected in sympatry, comprised of Deer, Goat and Man. Given that neither the number of hosts nor the accessibility to these hosts were a limiting factor in this particular sampling site, these results indicate that both An. messeae and An. atroparvus tend to be selectively zoophilic than An. daciae. The seeming indiscriminate feeding behaviour by An. daciae could be attributed to the response of An. daciae to various chemical cues from different host species including CO2 and fatty acids (Knols et al., 1997; Costantini et al., 1998). Alternatively it could also indicate an inability of An. daciae to discern the presence of more

114 than one host within a small area (Thomas 1951 cited in Port & Boreham, 1980) as observed in Pettits Animal Farm.

Interestingly, compared to An. messeae and An. atroparvus, 7 An. daciae females had fed on more than one host, an indication of an interruption in the bloodmeal. In fact, 15% of bloodfed mosquitoes screened had fed on multiple hosts. The prevalence of multiple bloodmeals has been reported in various studies conducted on biting and feeding behaviour of mosquitoes (Zimmerman et al., 2006; Michael et al., 2001; Gordon et al., 1991; Boreham & Lenahan, 1979). This detection of multiple hosts in a single bloodmeal is of particular interest in the transmission of arbo-diseases. A host-seeking female often imbibes a single uninterrupted (unmixed) bloodmeal, however, host irritability, incomplete feeding (Davies, 1990) or even a possible infection (viral or parasitic) of the vector (Rosomer et al., 1989) can lead to interrupted feeding. This then results in a female taking multiple meals, i.e., two or more feeds in a single gonotrophic cycle (Boreham & Garrett-Jones, 1973; Boreham, 1975; Rosomer et al., 1989). Interruption of feeding increases host-vector contact, thereby increasing the likelihood of pathogen transmission from the host to the mosquito and to other hosts. Thus the selection of multiple hosts such as birds (which serve as reservoirs for arbo diseases such as WNv and Sindbis) and man (through which arbodiseases such as malaria can be circulated among the population) observed in An. daciae in Norfolk as well as in Somerset presents it as a potential bridge vector of the West Nile virus.

Of equal interest is the feeding by An. daciae on exotic species such as R. tarandus (reindeer) and P. pennata (Darwin’s Rhea), which further illustrates the potential role that it could play as a bridge vector. The reindeer has been identified as a reservoir host for the enzootic Tahyna and Inkoo viruses (Brummer-Korvenkontio, 1973; L’vov et al., 1989) that commonly occur in Northern Europe (Ramsdale & Gunn, 2005). This unusual host choice by An. daciae (as well as An. messeae and Cs. subochrea) and that of local deer and man (An. daciae & An. messeae, herein; Cs. subochrea (Ramsdale & Snow, 1995) (Table 4.8), suggests that the importation of infected reindeer from other countries where these diseases are endemic, could pose a health risk to both human and wild deer populations in the UK. Screening measures should be undertaken to ensure animals are free of these diseases prior to importation.

115

Prior to modernisation of houses, An. messeae was found to rest indoors and was thought to be solely anthropophilic (Snow, 1990; Ramsdale & Gunn, 2005). Anopheles atroparvus was previously recorded to be voraciously anthropophilic and as such was circumstantially incriminated as a vector of malaria in the UK post WWI (Shute & Maryon, 1974; reviewed in Cranston et al., 1987 & Rees & Snow, 1990). However, since the eradication of malaria from the UK (Dobson, 1989), An. atroparvus has been recorded to be largely zoophilic (Cranston et al., 1987; Snow, 1990). Given the host selection reported herein, it is likely that An. atroparvus and An. messeae, as well as An. daciae could also serve as vector of Tahyna & Sindbis viruses and that An. messeae could play a role in the maintenance of arboviruses such as WNv within bird populations. The documented role of An. atroparvus in malaria transmission in the UK and the recent findings of two specimens feeding on a human host does not eliminate this species as a potential vector of human as malaria, as malaria is no longer endemic in the UK.

Host selection of Cx pipiens was unclear, in part because of the uncertainty of its taxonomic status (Chapter 3, section 3.1). In the UK, prior to our proposal of synonymy between Cx. pipiens f. pipiens and Cx. pipiens f. molestus, a distinction in feeding behaviour between the two forms was recorded: Cx. pipiens f. pipiens was ornithophilic and Cx. pipiens f. molestus was anthropophilic (Chapter 3, section 3.1). In the US (Apperson et al., 2002; Patrican et al., 2007) and southern France (Balenghien et al., 2006), Cx. pipiens has been recorded to feed on mammals and birds whereas in Sweden Culex pipiens is recorded to be primarily ornithophilic. Results presented herein show Cx. pipiens (in its current proposed synonymy; Chapter 3, Section 3.5) to be capable of feeding on both bird (Anglesey) and man (Somerset), albeit in two different localities. Its role as an incriminated vector of diseases, such as WNv in the US (Marfin et al., 2001) and Europe (Platonov et al., 2001; Estevez et al., 2005; Romi et al., 2004), Japanese Encephalitis in Asia (Johansen et al., 2002) as well as Rift Valley Fever in Africa (Turrell et al., 1996) and its current non distinctive feeding behaviour does implicate it as major candidate for the transmission of enzoonotic diseases in the UK.

Only a single bloodfed Cx. torrentium female was captured in this study and as such little can be said about its vector potential. Host preference exhibited by Cx. torrentium has reported to be solely ornithophilic (Snow, 1990; Medlock et al., 2005) and the only bloodfed Cx. torrentium (DNA 10454) collected here (in a pillbox in Godney, Somerset) was found to have had a multiple meal on a chaffinch and dog, thus the latter comprises a new host record 116 for Cx. torrentium in the UK. Interestingly, chaffinches have been thought to be a reservoir host for the Sindbis virus (J. Lundström pers comm.) of which Cx. torrentium is an incriminated vector (Lundström et al., 1990a, b). In the UK, although transmission of Sindbis virus is thought to occur (Buckley et al., 2003), no evidence has been shown linking Cx. torrentium as a vector. Considering all factors, Cx. torrentium could definitely play a role in maintaining the virus in the local bird population, thus enabling the transmission by another mosquito species, such as An. messeae or An. daciae which feed on birds, but also have a wider host selection.

The single Cq. richiardii bloodfed specimen analysed showed a preference for man. Previously, host preference of Cq. richiardii showed an affinity for mammals (Service, 1971; Snow, 1990), man (Ramsdale & Snow, 1995; Medlock et al., 2005) and birds (Snow, 1990). This diversity was also observed by Balenghien et al. (2006) in southern France, where specimens of Cq. richiardii were collected in both horse and bird baited traps and in human landing catches. Based on this diversity and the collection of specimens from near a West Nile virus case (Higgs et al., 2004), Cq. richiardii was considered to be a potential vector of WNv in the UK (Medlock et al., 2005; Higgs et al., 2004). However, its role as a potential vector, based on a single specimen, cannot be inferred in this study.

In general, host specific primers designed in this study could successfully identify six out of the seven target hosts however, there was still a need for sequencing in order to verify PCR products and to avoid the misidentification of hosts by preferential amplification of mosquito DNA. The use of universal CytB primers of Boakye et al, (1999) on field-caught samples showed these primers to identify arthropod blood meal and to also amplify mosquito DNA. Thus five host specific reverse and one host specific forward primer were designed to ensure amplification of target hosts. While host specific primers successfully amplified target host when control specimens were used, insect DNA was still preferentially amplified in 67% of individuals sequenced (See sections 4.4.2, 4.4.3). The seemingly ‘selective’ amplification of insect over host DNA could be attributed to the preservation of bloodfed mosquitoes. The preservation method of drying bloodfed mosquitoes does not allow for the extraction of vertebrate DNA only, thus enabling the competing amplification of mosquito DNA as shown in section 4.4.3. This problem could be overcome by the freeze killing of field caught bloodfed mosquitoes, dissecting the abdomen and smearing the blood meal on a piece of filter

117 paper, which is then stored in a container with silica (Weitz et al., 1956). Thus reducing the presence of insect DNA from the sample altogether.

The inability to consistently identify a host in bloodfed mosquitoes (both fully bloodfed and partially bloodfed) coupled with the lack of clear unambiguous sequences (4%) could lead to inaccurate identification of hosts, as illustrated by the detection of L. meridionalis in two individuals. This warrants the need for the further development of robust and sensitive molecular methods. One option would be to clone all fragments obtained with the use of universal CytB primers on fully bloodfed female mosquitoes. Cloned fragments could then be sequenced to determine host species selected which could in turn be used as a template on which more efficient primers could be designed. Alternatively, the use of Real- Time (RT) PCR has been optimised for the identification of mosquito blood meals (van den Hurk et al., 2007). The use of host-specific primers and probes allows for the quantification of DNA during the exponential amplification of the template (Heid et al., 1996), thereby determining the amount of host DNA present in a single mixed or unmixed blood meal. The primers designed herein could be used to in the design of a real-time PCR. Though RT-PCR eliminates the need for the visual identification of DNA fragments, it still requires very specific primers and verification by sequencing, as demonstrated here.

However, despite these shortcomings, this study was able to provide molecular evidence to the initial identification of potential vectors and bridge vectors in the UK, as well as identifying means through which arbo viruses such as WNv or even Sindbis could be transmitted at low levels (Buckley et al., 2003) in the UK. Unfortunately, the incrimination/implication of a vector cannot be done solely on the identification of host species using a mosquito blood meal. As demonstrated by Spielman (1986) and Beach et al. (1985), the release of saliva, particularly apyrase, during probing and feeding, can result in the release of the pathogen without taking a blood meal. Thus in addition to host selection, the isolation or detection of pathogen (either viral or parasitic) within the salivary glands of the mosquito will serve to incriminate species as vectors. Future work should focus on incrimination and determination of the role of British mosquitoes as primary and secondary vectors of disease, through viral and parasitic isolation and the collection of vectors in areas afflicted with viral or parasitic infection (Defoliart et al., 1987; Janini et al., 1995; Dutta et al., 1997). This knowledge would provide a better understanding of vector competency in

118

British mosquitoes, a tailored monitoring and surveillance system as well as the implementation of a targeted vector control programme.

119

Chapter 6

Occurrence and habitat preference of mosquitoes in southern England and northern Wales

120

6 Occurrence and Habitat Preference of Mosquitoes in southern England and northern Wales

6.1 Current occurence of British mosquitoes

As part of a wider study into malaria in England and Wales, Nutall et al. (1901) produced the first guide to the distribution of Anopheles mosquitoes, but did not differentiate between species. Lang (1918) reported occurrence of An. claviger (as An. bifurcatus), An. maculipennis s.l. and An. plumbeus in England and Wales and later Ashworth (1927) presented distribution maps for Scotland. Various species were reported in the works of Marshall (1938) and a review of the distribution of the mosquitoes of Ireland was added much later by Ashe et al. (1991). The first detailed distribution maps of UK mosquitoes were provided by Rees & Snow (1990, 1992, 1994, 1996) by genera: Anopheles (Rees & Snow, 1990; also reviewed by Snow, 1998), Culex (Rees & Snow, 1992), Coquillettidia, Culiseta and Orthopodomyia (Rees & Snow, 1994), Aedes and Dahliana (as Aedes, Aediomorphus and Finlaya) (Rees & Snow, 1995) and Ochlerotatus (as Aedes, subgenus Ochlerotatus) (Rees & Snow, 1996). Information gathered for these maps were obtained from the British Mosquito Recording Scheme and on the older records (Lang, 1918; 1920; Marshall, 1938) and, although the maps are highly useful, included no detailed information on component species in cryptic complexes, such as An. maculipennis s.l. In light of new molecular techniques to differentiate cryptic species (Proft et al., 1999; Linton et al., 2002a; Linton et al., 2005; herein Chapter 3), updated specific maps can and must now be provided, as this information is imperative in risk assessment and vector control strategies for mosquito-borne diseases in the UK in the future.

6.2 Aims The objectives of this study were:

[1] To document the species of mosquitoes collected in 7 counties in southern England and northern Wales, using integrated systematic methods for accurate species identification and

[2] To accurately characterise the preferred mosquito larval habitats and adult resting sites.

121

6.3 Materials and Methods

6.3.1 Collection of larvae and pupae

Larval and pupal stages were collected from natural (stream pools, stream margins, ponds) and artificial habitats (buckets, water troughs) using the dipping method outlined in Chapter 2 Section 2.2.1

Collected specimens were then individually reared to the adult stage. Emerged adults and associated larval and pupal skins were stored and labelled as detailed in Chapter 2, section 2.2.1.2.

6.3.2 Collection of resting and host seeking Adults

Both resting and host seeking adults were collected from brick shelters, war bunkers and animal shelters using manual aspirators and the Mosquito Magnet Trap Liberty Pro respectively (Chapter 2 Fig 2.2.2). Collected adult specimens were then processed according to the method detailed in Chapter 2 Section 2.2.2.

6.4 Results

Of the adult collections, 861 adults (resting, n=728; host seeking, n=133) were identified to species (Table 6.1). While 564 specimens were identified from the aquatic collections. In total of 13 species were collected: An. algeriensis, An. atroparvus, An. claviger, An. daciae, An. messeae, Cs. annulata, Cs. subochrea, Cq. richiardii, Cx. pipiens, Cx. torrentium, Da. geniculata, Oc. detritus and Oc. leucomelas (Table 6.1).

Human landing collections were not undertaken, but the single representative of Da. geniculata collected in this study was captured landing on (and presumably trying to bite) the author at dusk in the Bridestowe caravan park in Bridestowe, Devon.

122

Species Collection type Anglesey Caernarfonshire Devon Kent Norfolk Somerset Suffolk (n=233) (n=35) (n=340) (n=108) (n=565) (n=133) (n=12) An. algeriensis Aquatic 12 ------An. atroparvus Resting ------97 14 -- -- An. claviger Aquatic 22 -- 39 -- 1 -- -- Host seeking -- -- 5 -- 4 -- -- An. daciae Aquatic 1 -- -- 2 -- 3 Resting ------6 373 86 -- An. messeae Aquatic -- -- 2 ------3 Resting -- -- 97 5 13 7 -- Cq. richiardii Resting ------3 1 -- Host Seeking ------115 -- -- Cx. pipiens Aquatic 105 21 139 -- 27 24 6 Resting 1 -- 8 -- -- 8 -- Host Seeking 1 1 Cx. torrentium Aquatic 34 14 45 -- 12 -- -- Resting ------1 -- Cs. annulata Aquatic 41 -- 3 ------Resting 2 -- 1 -- -- 3 -- Host Seeking 2 Cs. subochrea Aquatic 7 ------Cs. subochrea Resting ------3 -- Da. geniculata Human Landing -- -- 1 ------Oc. detritus Host Seeking 4 ------Oc. leucomelas Host Seeking 2 ------

Table 6.1 Numbers of larval and adults collected per species in each county, by collection type. Immatures (n=222), resting (n=3) and 8 host-seeking individuals were collected using the Mosquito Magnet® trap in Anglesey. Human landing (n=1), larvae/pupae (n=208) resting (n=106) and host seeking female (n=6) collections were made in Devon. Only Resting (n=108) adults were identified in Kent, Larvae/pupae (n=42), resting (n=403) and host-seeking (n=220) adult collections were made in Norfolk. Only Larvae/pupae (n=24) and resting (n=109) collections were made in Somerset. Larvae/pupae (n=12) were collected in Suffolk and in Wales (n=35).

123

6.4.1 Characterisation of aquatic stages

Of the 60 individual aquatic localities sampled, twenty-four mixed species collections were found, with single species (possibly isofamilies) being detected in the remainder. Even in mixed species collections, one species tended to be overwhelmingly dominant, with the others most frequently represented by only a few specimens.

The majority of immature collections came from Devon, comprising 43% of all specimens collected, followed by Anglesey (n=222). Of all immatures collected, 60% (n=956) belonged to genus Culex, which were found in all counties except Kent (Figure 6.1).

In general, immatures of Culex spp. could be found in artificial habitats (84%) such as discarded bath tubs, boats and buckets, generally lacking any aquatic vegetation although some specimens were collected in natural habitats such as stream pools, groundpools and ditches containing floating and emergent vegetation (Figure 6.1). Anopheles larvae, however, were found in 8.9% of all larval collections showing a strong preference for natural habitats, such as ground pools, stream margins and ditches (Figure 6.2), in which moderate amounts of floating and emergent vegetation were present (Figure 6.1). However a few individuals were collected in artificial habitats, including: a tyre in Anglesey (n=2), a fire bucket in Bridestowe Caravan Park and in a water trough in Caernarfonshire, Wales (Figure 6.2). Aquatic stages of the genus Culiseta were found only in two counties (Table 6.2) in natural habitats including stream margins and stream pools, with 0.3% of all Culiseta collected also being detected in moored boats with collections of rainwater. Interestingly, buckets were found to be the most productive artificial habitat, yielding 17% of all immature habitats sampled, while ditches and stream pools were the most productive natural habitats (10% each) (Figure 6.3).

124

Figure 6.1 Graph showing the relative proportions of immature stages collected by genera in relation to the amount and type of vegetation found in the larval habitats. (Em = emergent vegetation; Fl = floating vegetation; All = floating and emergent vegetation).

125

Figure 6.2 Graph showing association of three mosquito genera (Anopheles, Culex and Culiseta) with different larval habitats in five regions of England: Devon (Dev), Norfolk (Nor), Somerset (Som), Suffolk (Suf) and two regions of northern Wales (Anglesey (Ang) and Caernarfonshire (Wales).

126

Figure 6.3 Relative proportions of mosquito immatures collected in July 2006 by habitat type. Natural habitats (denoted with ‘N’) yielded only 33% of total mosquitoes collected (n=530) and included stream margins, stream pools, ponds, lakes, ground pools and ditches. A total of 1072 specimens were collected in man-made habitats. Artificial habitats included water collections in forklift buckets, tarpaulin sheets, tyres, boats, buckets and troughs.

127 6.4.2 Collected British mosquitoes detailed by genera

6.4.2.1 Genus Anopheles

Five species of Anopheles were collected in this study: An. algeriensis, An. claviger and three species in the Maculipennis Group – An. atroparvus, An. daciae and An. messeae. Specimens of the Maculipennis group were identified using the ITS2 PCR_RFLP assay designed in Chapter 2. Despite searches in treeholes, where available, no Anopheles plumbeus were collected in this study.

6.4.2.1.1 Anopheles algeriensis

Anopheles algeriensis (n=12) was collected as immatures in three sites: two ditches (Larvae (L=3), Pupa (P=4)) and one ground pool (L=3, P=2), all with abundant vegetation, in the calcarious marshland habitat of Cors Goch Nature Reserve in Anglesey. No adults were collected in this study (Figure 6.4).

6.4.2.1.2 Anopheles atroparvus

Five adult An. atroparvus (four non-bloodfed females, one male) were collected from a brick shelter on Salt Lane in the Cliffe Marshes in Kent were collected together with An. messeae and an unidentified Culiseta. In another bunker, used mainly as a sheep corral, also located in the Cliffe marshes, An. atroparvus (n=92; 26 bloodfed females, 62 non-bloodfed females and four males) was collected in sympatry with An. daciae and An. messeae. Fourteen adult An. atroparvus were collected resting in goat houses (n= 8) and reindeer stable (n= 6) in Pettits Animal Farm in Norfolk, sympatrically resting with An. daciae and An. messeae. All but one of the females collected were bloodfed. No immature stages of An. atroparvus were collected in this study (Figure 6.4).

128

Figure 6.4 Occurrence map of 5 species of British Anopheles mosquitoes collected in southern England and northern Wales in July and August 2006. Symbols contained within a circle indicate several collections in the locality indicated. Numbers written in the symbols indicate the total number of specimens collected. Co-ordinates for all collections are available in Chapter 2 Table 2.2. Anopheles algeriensis was collected as larvae in Anglesey; An. atroparvus as resting adults in Norfolk and Kent; An. claviger collected as larvae in Anglesey, Devon and Norfolk and as host seeking adults in Devon and Norfolk; An. daciae collected as larvae in Anglesey, Norfolk and Suffolk and as resting adults in Somerset, Kent and Norfolk; and An. messeae collected as larvae in Devon and Suffolk; as resting adults in Somerset, Kent and Norfolk. 129

6.4.2.1.3 Anopheles claviger

In Devon, An. claviger was detected in three sites: in a stream near Yelland farm (L=1, P=2), in a woodland pool near Boating World (L=12, P=2) and a stream pool (L=3, P=3) in Simmon’s Park. It was also collected in three sites in the Exminster Nature Reserve, Devon; in a ditch (L=5, P=4) and in two stream margins (L=5, P=1). One individual was collected in a stream margin on Horsey Road, in Reedham, Norfolk. Aquatic stages were found in three sites in Anglesey: a single An. claviger pupa was collected in a stream pool near a horse farm, and larvae and pupae were collected in a stream near Mallraeth Marsh (L=8, P=7) and in a ditch near a cottage in Benllech (L=4, P=2). Anopheles claviger was most often found as pure immature collections, with the exceptions of only three collections. The species was collected sympatrically with An. messeae and Cs. annulata (in a stream margin in Exminster marshes, Devon), with An. daciae (in a stream margin in Norfolk and with Cx. torrentium (in a ditch in Benllech, Anglesey).

Five host seeking adults were collected near horse stables in the Miniature Pony Centre in Moretonhampstead, Devon and four in Pettits animal farm in Reedham, Norfolk, where miniature donkeys, miniature horses, reindeer, Darwin’s Rhea and goats were present (Figure 6.4).

6.4.2.1.4 Anopheles daciae

A single An. daciae pupa was collected in Anglesey, in a stream pool behind a horse farm. In Norfolk, two larvae were detected in a stream margin along Horsey Road. Larvae (n=3) of An. daciae were also found in two ditches in the Carlton Marsh Nature Reserve in Suffolk, in sympatry with An. messeae.

All resting collections made in Norfolk came from animal stables in Pettits Animal Farm: in goat houses (n=197), miniature horse and donkey stables (n=45), reindeer stables (n=124) and a Rhea stable (n=7). Of these, 72.6% (n=271) were bloodfed (see Chapter 5). Sympatry with An. messeae was observed in all the animal stables sampled. In addition, all three species of the Maculipennis Group were collected in both goat and reindeer stables in Norfolk. Eighty-six adults were collected from three localities in Godney, Somerset: a horse 130 stable (n=7) and two pillboxes (n=79). Anopheles daciae adults collected in the horse stable also was found resting with An. messeae and Cq. richiardii. In the first pillbox in Godney village, An. daciae was found resting in sympatry with An. messeae, Cx. pipiens, Cx. torrentium, Cs. annulata and Cs. subochrea and with An. messeae, Cx. pipiens and Cs. annulata in the second pillbox (Figure 6.4).

6.4.2.1.5 Anopheles messeae

Anopheles messeae larvae were found in pure populations in two stream margins in Exminster Nature Reserve, Devon. It was also collected in two ditches in the Carlton Marshes Nature Reserve, in Suffolk, together with An. daciae.

In the Cliffe Marshes in Kent, An. messeae was collected in sympatry with An. daciae and An. atroparvus in an abandoned brick shelter and together with An. atroparvus in a bunker on Salt Lane (n=1). Anopheles messeae was also found in Norfolk in goat stables (n=1), miniature horse and donkey stables (n=4), reindeer stable (n=7), in sympatry with An. daciae as well as An. atroparvus and in a Rhea stable (n=1). Ten individuals were found to have had a blood meal (see Chapter 5). Two females were resting in a horse stable in Somerset, one in a pillbox on near Godney Farm and five individuals were resting in a pillbox on Garslade farm (Figure 6.4).

131

6.4.2.2 Genus Culex

6.4.2.2.1 Culex pipiens

In Anglesey, fifteen localities were positive for Cx. pipiens (13 artificial and 2 natural sites) in Benllech (n=36), Brynteg (n=20), Llangferfechan (n=12), Newborough (n=6), Penraeth (n=26) and Mallraeth (n=5) (Table 5.3). In Devon, Cx. pipiens was detected in 14 localities (11 artificial, 3 natural) in Belstone (n=28), Bridestowe (n= 27), Cheriton Bishop CP (n=32), Moretonhampstead (n=40), St Erney (n=1), Trevollard (n=9) and Dartmoor forest (n=2). Culex pipiens were found in two separate boats in Norfolk in Ranworth Wildlife Centre (n=11) and in the margins of a small stream in West Somerton (n=16). All three habitats sampled in Somerset were natural: a ground pool on Garslade farm (n= 3) and two ditches (n= 13, n= 8). The same was found in Suffolk, where all aquatic stages of Cx. pipiens were collected in natural habitats in and around Beccles Marsh. Cx. pipiens (L=8, P=13) was also detected near horse stables near Betws-y-Coed in Caernarfonshire, in artificial habitats including horse troughs and barbeque bowls.

One adult Cx. pipiens was collected resting in a shed in Bird World, Anglesey, in sympatry with An. maculipennis s.l. and Cs. annulata. All seven individuals in Devon were collected resting inside a horse stable in Eastland Horse Farm; none was blood fed. A single host-seeking female were collected using the Mosquito Magnet Trap ®, near a chicken coop in the Miniature Pony Centre in the New Forest and one resting adult was collected in sympatry with An. messeae and Cs. annulata in a brick shelter in Exminster Nature Reserve, also Devon. Culex pipiens was collected as resting adults in two pillboxes in Godney, Somerset (n=8) in sympatry with An. daciae, An. messeae and Cs. subochrea (Figure 6.5).

6.4.2.2.2 Culex torrentium

Culex torrentium was collected in 11 habitats (9 artificial and 2 natural) in four localities in Anglesey: Benllech (L=9, P=3), Brynteg (L=3, P=5), Llangferfechan (L=4, P=7) and Penraeth (P= 3). Ten habitats (8 artificial and 2 natural) in six localities sampled in Devon were also found to support Cx. torrentium: Belstone (L=3, P=9), Bridestowe (L=15, P=8), Cheriton Bishop (L=1, P=2), Moretonhampstead (L=1, P=2) and Dartmoor Forest (P= 3) and 132

St. Erney (P=1). In Norfolk, a total of twelve individuals (L=1, P=11) were collected in a moored boat in Ranworth Wildlife Centre. In Caernarfonshire, Cx. torrentium was found in three artificial habitats in Ty Coch Riding Stables (L=11, P=3) near Betws-y-Coed. A single bloodfed adult was collected resting in a pillbox in Godney village in Somerset (Figure 6.5). Culex pipiens and Cx. torrentium were collected in sympatry in 68.6% (n=24) of all localities sampled: Anglesey (n=10), Devon (n=10), Norfolk (n=1) and Wales (n=3).

133

Figure 6.5 Presence of Culex mosquitoes collected in northern Wales and in southern England in this study. Symbols contained within a larger circle indicate several collections in the specified locality. Numbers written in the symbols indicate the total number of specimens collected. Co-ordinates for all collections are available in Chapter 2 Table 2.2. Culex pipiens was collected as larvae in Anglesey, Devon, Norfolk, Somerset, Suffolk and in northern Wales and as resting adults in Anglesey and Somerset; as host seeking adults in Devon. Culex torrentium collected as larvae in Devon, Norfolk, England, in Anglesey and Caernarfonshire in Wales and as resting adults in Somerset.

134

6.4.2.3 Genus Culiseta

Immature stages of the genus Culiseta were relatively uncommon, with only ninety-six individuals collected in this study. Of these, only sixty were identified to species Table 6.1. Both Culiseta species were collected in sympatry as larvae in a stream pool in Cronllech Manor Farm, Anglesey and as resting adults in Somerset.

6.4.2.3.1 Culiseta annulata

Forty individuals (L=5, P=35) of Cs. annulata were found in a stream pool in Cronllech Manor Farm in Anglesey together with immature An. claviger, Cx. pipiens and Cs. subochrea. A single collection of Cs. annulata was collected in an abandoned tyre in the Mallraeth Marshes in Anglesey together with Cx. pipiens and a single undetermined Anopheles. Culiseta annulata was also collected in a stream margin in Exminster Nature Reserve, Devon together with An. claviger. Resting adults were collected in three localities (Table 6.1); none of the females were bloodfed. Two individuals were collected in shed in Bird World on Anglesey together with An. maculipennis s.l. and Culex spp. (Figure 6.6)

6.4.2.3.2 Culiseta subochrea

Culiseta subochrea (n=7) was collected in a stream pool in Cronllech Manor Farm. This was the only larval collection from which Cs. subochrea was identified. Resting adult females of Cs. subochrea (n=3) were collected in a pillbox in Godney village Somerset; two of these were blood fed (Figure 6.6).

135

Figure 6.6 Occurrence map of Culiseta mosquitoes collected in southern England and northern Wales in July and August 2006. Symbols contained in a larger circle indicate several collections within the specified locality. Numbers written in the symbols indicate the total number of specimens collected. Co-ordinates for all collections are available in Chapter 2 Table 2.2. Culiseta annulata collected as larvae in Anglesey; as resting adults in Devon and Somerset and Cs. subochrea collected as larvae in Anglesey; as resting adults in Somerset. 136

6.4.2.4 Genus Coquillettidia

Host seeking females were collected in Pettits Animal Farm in Norfolk (Table 6.1), where further specimens were collected resting in the miniature horse and donkey stables (n=1) and in goat stables (n=2). A single resting adult female was collected in a horse stable on Garslade farm, Godney, Somerset (Figure 6.7).

6.4.2.5 Genus Dahliana

A single Da. geniculata female was manually collected from the knee of the author (presumably looking for a host) in Bridestowe Caravan Park, Devon (Figure 6.7). Despite searching in tree holes, where possible, no larval stages, or resting adults, were collected.

6.4.2.6 Genus Ochlerotatus

Both Oc. detritus and Oc. leucomelas (Figure 6.7) were collected as host seeking females (Table 6.1), in the environs of Bird World in Anglesey. Sampling of aquatic stages within the vicinity was fruitless, despite extensive sampling of nearby water sources.

137

Figure 6.7 Occurrence map of Coquillettidia, Dahliana and Ochlerotatus mosquitoes collected in southern England and northern Wales in July 2006. Symbols contained in a larger circle indicate several collections within the specified locality. Numbers written in the symbols indicate the total number of specimens collected. Co-ordinates for all collections are available in Chapter 2 Table 2.2. Coquillettidia richiardii was collected as resting adults in Somerset and as host seeking adults in Norfolk. Dahliana geniculata was collected as a host-seeking adult in Devon. Ochlerotatus detritus and Oc. leucomelas were collected only as host seeking adults in Anglesey. 138

6.5 Discussion

This study, although somewhat comprehensive across seven counties in southern England and Wales, comprises only a “snap shot” of the composition and relative abundance of mosquitoes in the UK, as it reflects only mosquitoes present in July (Anglesey, Caernarfonshire, Devon, Norfolk, Somerset & Suffolk) and August (Kent) of 2006. Of the 33 documented British mosquito species, only 13 species, belonging to six genera (Anopheles, Coquillettidia, Culex, Culiseta, Dahliana and Ochlerotatus), were collected in this study as follows: An. algeriensis, An. atroparvus, An. claviger, An. daciae, An. messeae, Cq. richiardii, Cs. annulata, Cs. subochrea, Cx. pipiens, Cx. torrentium, Da. geniculata, Oc. detritus and Oc. leucomelas. All these species have previously been reported in the UK (Cranston et al., 1987; Snow, 1990; Linton et al., 2002a; Linton et al., 2005). Adults from twenty-two species have been documented (Cranston et al., 1987; Snow, 1990; Snow & Medlock, 2008) from May to September. Of these, 13 were not collected in this study: Ae. cinereus, Ae. geminus, Ae. vexans, An. plumbeus, Cs. morsitans, Oc. annulipes, Oc. cantans, Oc. caspius, Oc. dorsalis, Oc. flavescens, Oc. punctor, Oc. rusticus and Or. pulcripalpis.

The absence of species such as Oc. flavescens could be attributed to its recorded patchy/sparse distribution (Rees & Snow, 1996). This would also explain the absence of species such as Oc. communis (reported four times, Snow et al., 1998), Oc. sticticus (last reported in 1938, Rees & Snow, 1996), Cx. territans, Cx. modestus (last reported in Hayling Island in 1945, Snow et al., 1998), Culiseta longiareolata (recorded on three occasions, Rees & Snow, 1994), Cs. litorea (last reported in 1955 in Surrey, Rees & Snow, 1994) and Cs. alaskaensis (Linton et al., 2005 confirmed one specimen of this from Kent). The lack of recent records suggests that these species are either elusive or may no longer occur in England. While Ae. cinereus, An. plumbeus, Cs. morsitans, Oc. annulipes, Oc. caspius, Oc. dorsalis, Oc. punctor and Oc. rusticus were not collected in this study, recent records of these species were made in Epping Forest (Hutchinson et al., 2007; Snow & Medlock, 2008), in the Isle of Sheppey (Hutchinson et al., 2007) and in Wicken Fen (Hutchinson et al., 2007). Given that Aedes and Ochlerotatus species overwinter as eggs (Snow, 1990), the dry summer of 2006 could account for the lack of suitable aquatic habitats thus delaying the emergence of adults.

139

In England, a total of ten species were collected from Devon (An. claviger, An. messeae, Cx. pipiens, Cx. torrentium, Cs. annulata and Dahliana geniculata), Kent (An. atroparvus, An. daciae, An. messeae, Cs. annulata), Norfolk (An. atroparvus, An. claviger, An. daciae, An. messeae, Cq. richiardii, Cx. pipiens and Cx. torrentium) Somerset (An. daciae, An. messeae, Cq. richiardii, Cs. annulata, Cs. subochrea, Cx. pipiens and Cx. torrentium) and Suffolk (An. daciae, An. messeae and Cx. pipiens). The presence of An. atroparvus in Norfolk, An. messeae in Devon, An. daciae in Kent, Norfolk, Somerset and Suffolk, Cs. subochrea in Somerset and Cx. torrentium in Somerset comprise new distribution records in England.

The Welsh island of Anglesey proved highly speciose, with 9 of the 13 taxa discovered in the whole survey: An. algeriensis, An. claviger, An. daciae, Cx. pipiens, Cx. torrentium, Cs. annulata, Cs. subochrea, Oc. detritus and Oc. leucomelas. Only two species were collected in Caernarfonshire, Wales: Cx. pipiens and Cx. torrentium. Given the lack of detailed mosquito surveys in Wales (Morgan, 1978; Rees & Rees, 1989), this study contributes to the reported diversity in this region of UK in particular. The collection of An. daciae, Cs. subochrea and Oc. leucomelas, in Anglesey, as well as Cx. torrentium in Caernarfonshire and Anglesey comprise new country and distribution records.

Most surprisingly, this study showed that Anopheles daciae was one of the most widespread and locally dominant species in this study, despite its relatively recent discovery (Linton et al., 2005). Prior to this study, little data were available on either the ecology of immatures and adults. The large collections of An. maculipennis s.l. herein provided sufficient material for the development and optimisation of an ITS2 PCR-RFLP assay (see Chapter 3) that discriminates between the three members of the Maculipennis Group occurring in England. Anopheles daciae was identified in the counties of Devon, Norfolk, Suffolk, Somerset (reconfirming its presence) and in Anglesey, Wales (Chapter 3). The Welsh record of An. daciae near Mallraeth is the most northerly documentation of An. daciae thus far. Resting adults were found in sympatry with An. messeae as well as An. atroparvus, whilst immatures were found in sympatry with An. messeae in Suffolk. This sympatric larval occurrence was also documented the type series in Romania, where An. daciae adults were found with An. atroparvus (Nicolescu et al., 2004). That An. daciae, the newest member of the British mosquito fauna is actually one of the most common and numerous species in the UK collections, serves to highlight how little we understand of the current demographics of local mosquito species in the UK. 140

Also found in this study was three relatively uncommon species: An. algeriensis, Cs. subochrea and Oc. leucomelas. Anopheles algeriensis, a predominantly Mediterranean species, was recorded in Norfolk by Edwards (1932) and later reconfirmed some 20 years later by Hart (1954). The presence of Anopheles algeriensis in the Cors Gogh Nature Reserve in Anglesey, herein, confirms earlier reports of Morgan (1987) and Rees & Rees (1989), although, no samples were detected in Norfolk in this survey.

Culiseta subochrea was previously reported to occur only in southeast England (Marshall, 1938; Rees & Snow, 1994; Snow et al., 1998). In the present study, larvae were collected in Anglesey in sympatry with Cs. annulata (as also reported by Cranston et al., 1987) and resting adults were collected in Somerset. It has more recently been reported in Epping Forest, Essex (Snow & Medlock, 2008). Both collections make these the most recent records of this species in the literature since 1968 (Cranston et al., 1987).

Only two host-seeking adults of Ochlerotatus leucomelas were collected in this study, from Bird World in Anglesey. The only other occurrence records for this species, in the UK, were made by Carr (1919, in Marshall, 1938) near Nottingham and by Martini (1920, in Cranston et al., 1987) in Dartford, Kent. This collection in Anglesey reported here constitutes the most current occurrence in the UK and also a new distribution record in Anglesey for Oc. leucomelas. Although manual collections of resting adults were carried out in Bird World, no specimens of Oc. leucomelas were found indoors, which suggests that it may rest outdoors on vegetation.

In general it was found that aquatic stages of Anopheles and the majority of Culiseta larvae were most closely associated with natural environments, such as pools, streams and ditches, with floating (e.g. algae) and/or emergent (e.g. reeds, grasses) vegetation. In contrast, Culex mosquitoes were found to favour artificial habitats (e.g. buckets, bath tubs, water troughs). This observation was also reported by Marshall (1938), Rees and Snow (1990), Rees and Snow (1992) and Rees and` Snow (1994). However, contrary to reports of Cranston et al. (1987), who documented the presence of An. claviger in both brackish as well as freshwater, An. claviger was only found in freshwater habitats, in this study, such as pools, ditches and streams. In addition, 7 specimens of Anopheles sp. were collected in a tyre (Anglesey) or water troughs (Anglesey, Devon). Given that these specimens were preserved in 80% ethanol, 141 possible misidentification of the genus could have occurred, thus accounting for the ecological anomaly. Interestingly, Cx. torrentium, is reported to “exhibit a marked preference” for artificial habitats (Cranston et al., 1987; Snow, 1990) and it was thought that with the increase in suitable contained-habitats Cx. torrentium could be replacing Cx. pipiens (Mattingly, 1967). In this study, Cx. torrentium was commonly collected in artificial habitats with the exception of two sites, where it was collected in a ditch (Anglesey) and a ground pool (Devon). However, when collected in sympatry with Cx. pipiens, Cx. torrentium was less numerous.

Resting adult collections carried out in animal shelters, bunkers and pillboxes, recovered 8 species, of which members of the Maculipennis Complex were by far the most abundant. Seemingly localised abundance of An. atroparvus in Kent, An. daciae in Norfolk and Somerset and An. messeae in Devon was observed, suggesting that this method of capturing adult Anopheles, which also allow studies of natural infection and blood meal analysis (see Chapter 5), is effective for malarial surveillance in the UK. However collecting indoor resting adults in this manner alone excludes the collection of species like An. algeriensis, Cq. richiardii, Cs. morsitans, Oc. cantans, Oc. caspius and Oc. rusticus which prefer to rest in vegetation outdoors (Cranston et al., 1997; Snow, 1990), thus presenting a bias in the sampling strategy and reducing the diversity of adult species collected in each sampling site.

One solution to overcome this sampling bias was to place the Mosquito Magnet Trap® in areas where resting adults were collected as well immature collections made. With this trap, 5 species were collected as host-seeking adults in addition to either immature or resting adult collections: An. claviger, An. maculipennis s.l., Cq. richiardii, Cs. annulata, Cx. pipiens. Only two species of Ochlerotatus (Oc. detritus and Oc. leucomelas) were collected solely as host seeking adults in Anglesey despite attempts at collecting resting adults and immatures. This illustrates the importance of using different sampling methods to ascertain the occurrence of species in an area. The use of this trap as a tool for monitoring species diversity was also reported by Hutchinson et al. (2007) who compared mosquito collections using a CDC Light Trap with those of the Mosquito Magnet Trap ® in England (Chadwell Heath, Epping Forest, Isle of Sheppey and Wicken Fen). They found that the Mosquito Magnet® attracted a wider diversity of mosquito species such as An. claviger, Cq. richiardii, Oc. annulipes, Oc, cantans, Oc. caspius, Oc. geniculatus, Oc. punctor and Oc. rusticus compared to the CDC light traps. The combination of carbon dioxide and 1-Octen-3-ol (Octenol) in the Mosquito Magnet® trap was found to be attractive to host-seeking mosquito species (Takken & Kline, 1989; Becker et 142 al., 1995). Although, diversity of species recorded is affected by the number of Mosquito Magnet Traps ® used and localities in which they are placed (Brown et al., 2008). Thus, in wider arbovirus surveillance studies the use of: more than one Mosquito Magnet Trap ® in an area, sweep net and automated suction pootering of vegetation would perhaps be more effective in capturing various species for vector incrimination of other mosquito-borne diseases.

Although this field study only documents British mosquitoes across a limited geographic area, the ecology of both adults and aquatic stages of the 13 species collected have been characterised and recent knowledge of host preference determined. Data from this study have been accessioned into Mosquito Map (www.wrbu.org) and Mosquito Watch (http://www.cieh.org/policy/npap_uk_sightings.html) to serve as baseline data for future studies. With the resurgence of emerging (Chikungunya) and re-emerging (West Nile fever, Dengue fever, malaria) diseases in and around Europe, knowledge of local mosquito species is essential for the identification and incrimination of mosquito vectors. The need for further extensive field surveys across the season, using the combined collections of the Mosquito Magnet Trap®, resting indoor and outdoor adult resting collections and immature samplings, combined with revisiting former collection localities of rare taxa and establishing new ones is apparent.

143

Chapter 7

General Discussion

144

7 General Discussion

7.1 Discussion

Mosquitoes in the UK have been studied intermittently since the early 20th century, driven primarily by the need to understand vector-host relationships and the dynamics of malaria transmission (Nutall, 1901; Lang, 1918, 1920). Following this early work, information on local mosquito fauna increased and detailed publications on life cycle and morphology (Edwards, 1932), oviposition and host selection (Service, 1969; 1971) and identification keys (Cranston et al., 1987; Snow, 1990) were made available. Unfortunately, since eradication of malaria from the UK after WWII (Dobson, 1989), there has been a paucity of mosquito studies, leading to a significant gap in knowledge of the current ecology and distribution of British mosquitoes. This in turn makes it difficult to monitor, even predict, the introduction of enzoonotic diseases such as Chikungunya and West Nile virus into the UK. The importance of knowing the species present in a given region as well as the ability of these species to transmit emerging/re-emerging diseases, created the need to reassess the vector potential of mosquitoes endemic to the UK. With this intention, the overall aim of this thesis was to document and characterise the ecology (adult and larval habitats and host preference) of British mosquitoes so as to facilitate the identification of potential vector species.

Out of the 33 species recorded in the UK, 13 were collected in this study. Taking together ecological studies conducted by Snow & Medlock (2008) and Hutchinson et al. (2007), the continued presence of 24 species is currently documented in southern England. Species that remain elusive include Oc. communis, Oc. sticticus, Cx. territans, Cx. modestus, Culiseta longiareolata, Cs. litorea and Cs. alaskaensis suggesting either their possible absence or localised presence in the UK (Cranston et al., 1987). The two commonly collected species, An. daciae and Cx. pipiens, both belong to complex groups (Maculipennis Group and Pipiens Group respectively) and the accurate identification of members in a complex species is particularly important for studies on vector competency. Out of the 11 recognised members of the Palaearctic Maculipennis Group, 3 are considered to be efficient vectors of malaria (Chapter 2, section 2.1.3), of which An. atroparvus is one.

145

The presence of An. atroparvus (Snow et al., 1998) has been documented in the UK, together with An. messeae and An. daciae (Linton et al., 2002a; Linton et al., 2005). Yet prior to DNA sequencing of the nuclear ITS2 region (Linton et al., 2002a; Linton et al., 2005), which led to the development of the ITS2-PCR assay (Chapter 3, section 3.3.4), adult females of the Maculipennis Complex could not be reliably differentiated on morphology alone. In fact, recent publications have referred to both these taxa as “An. messeae” with the acknowledgement that An. daciae may well be included (Hutchinson et al., 2007; Snow & Medlock, 2008), or An. maculipennis s.l. The ITS2-PCR RFLP assay designed herein (Chapter 3) successfully distinguished all three species of the Maculipennis Group occurring in the UK. Despite only a 5-bp difference in the ITS2 region between An. daciae and A. messeae, the RFLP was able to accurately identify the two sibling species in all localities sampled in the UK. The most surprising result obtained from this assay was the prevalence of An. daciae, comprising 63.7% of the adult collections. Anopheles daciae was initially detected in 5 individuals collected Somerset in 2001 (Linton et al., 2005) and it was the latest addition to the British faunal list. Collections made in this study showed An. daciae to be present in not only in Somerset but in Norfolk, Kent, Suffolk and Anglesey as well. Interestingly, the presence of An. daciae in the UK corroborates the hypothesis of Edwards (1936), who suggested the presence of a third member of the Maculipennis Group in the UK. However it is more likely that the third member could have been An. daciae and not An. maculipennis s.s. as originally proposed.

While molecular markers such as ITS2 and COI are able to aid the differentiation of members of a species complex, this was not the case for the two forms of the Pipiens Complex studied here: Culex pipiens f. pipiens and Cx. pipiens f. molestus. Both forms, reported to occur in the UK, are morphologically and genetically indistinct with only differences in host and ecological preferences (Chapter 4, Section 4.1.2) to serve as a guide in discriminating between the two. Two assays, recently developed to differentiate the two forms, both failed here to consistently separate Cx. pipiens f. pipiens from Cx. pipiens f. molestus in the UK (Chapter 4, Section 4.4.3). In fact, the microsatellite assay (CQ11, Bahnck & Fonseca, 2006), detected Cx. pipiens f. molestus as well as hybrids of Cx. pipiens f. pipiens and Cx. pipiens f. molestus occurring aboveground (Chapter 4, Section 4.4.2), neither of which have been documented in the UK. Subsequent sequencing of a subset of the CQ11-identified individuals allowed for a comparison with the COI-RFLP assay of Shaikevich (2007); designed to differentiate both forms of the Pipiens Complex as well as Cx. torrentium. Only Cx. pipiens f. 146 pipiens and Cx. torrentium were detected with COI. Interestingly, 16 individuals identified by the CQ11 assay as either Cx. pipiens f. molestus, Cx. pipiens f. pipiens or hybrids were Cx. torrentium according to COI (Chapter 4, Figure 4.5). Dspite morphologically identifying all specimens prior to using both assays, the morphological misidentification of Cx. torrentium as Cx. pipiens s.l. resulted in the inaccurate detection of Cx. pipiens f. molestus and of hybrids, thus highlighting the unreliability of using the absence of pre-alar scales as a means to identify females of Cx. pipiens s.l. from Cx. torrentium (Section 4.5). In addition, the COI-RFLP assay had also identified Cx. pipiens f. molestus colony material as Cx. pipiens f. pipiens suggesting that either the colony was not Cx. pipiens f. molestus or that the polymorphisms on which the COI-RFLP, as well as the microsatellite assay, was designed cannot be used outside its geographical range. Further analysis of the mitochondrial marker showed a low variability between Cx. pipiens f. pipiens and Cx. pipiens f. molestus (Section 4.4.3, Figure 4.5), suggesting that the two forms of Culex pipiens may not be taxonomically viable.

The stark difference observed in the analysis of the Maculipennis Group and Cx. pipiens complex, emphasizes the need for a multi-characteristic support on the taxonomic differentiation of species prior to the development of molecular assays (DeSalle et al., 2005); for e.g. either by both morphology and genetic characteristics, or geographical and morphological characteristics. The differentiation of two putative forms of Cx. pipiens (Cx. pipiens f. pipiens and Cx. pipiens f. molestus) was supported primarily on the presence/absence of autogeny (ecology), purported differences in host selection (ecology) and morphology. In turn, independently designed molecular assays (genetic), using both microsatellite (CQII) and mitochondrial COI markers, alluded to the accurate identification of the two forms. Thus according to the proposition of DeSalle et al. (2005), there were three characteristics supporting the two forms of Cx. pipiens. Inconsistent results produced by both assays (Chapter 4, section 4.4.3) coupled with the indistinct morphological characteristics (Harbach et al., 1985) and variability of expressed autogeny in other mosquito species suggested, instead, that Cx. pipiens was a single species (Chapter 4, section 4.5). On the other hand, An. daciae, a recently recorded species of the Maculipennis Group in the UK, was first discriminated from An. messeae based on egg morphology as well as fixed polymorphisms on both nuclear ITS2 and COI markers (Nicolescu et al., 2004). The PCR-ITS2 assay developed in this study consistently and accurately differentiated An. daciae and An. messeae across their range including the UK, (Chapter 3, section 3.4.2), Romania, as well as in Poland and

147

Bulgaria (Linton, unpublished). Thus the presence of both morphological and genetic characteristics supports the species of the Maculipennis Group in the UK.

The final aspect of this study dealt with the host selection of field-caught bloodfed mosquitoes. The selection of hosts by a mosquito is an important determinant of the potential role played by the mosquito species in disease transmission. Up until now, no current data were available on the host selection of British mosquitoes, and therefore it is difficult to identify potential vectors and potential bridge vectors in the UK. Using individually designed primers for a fragment of mitochondrial CytB, mosquito bloodmeals were screened for six vertebrate hosts: Bird, Dog, Deer, Goat, Man and Horse.

The most interesting result here was the indiscriminate feeding behaviour observed in An. daciae. Its selection of non-endemic species such as the Reindeer and Darwin’s Rhea (both found in Pettits Animal Farm, Norfolk) and selection of more than one host in the same reproductive cycle is suggestive of An. daciae as a potential bridge vector. Having been found to feed on both birds and man opens the possibility for it to be a transmitter of Sindbis or even WNv, as antibodies for both have been detected in chicks in the UK (Buckley et al., 2003). The occurrence of An. daciae in Norfolk and Kent (chapter 3) and its observed selection of man (chapter 5), combined with the historical incidence of malaria in these counties (chapter 3, section 3.1) could imply its potential to transmit human diseases such as malaria. Aside from An. daciae, An. atroparvus, An. messeae were also found to have fed on more than one host in the same reproductive cycle. However, in comparison to the diverse selection of hosts by An. daciae, both An. atroparvus and An. messeae, seemed to be zoophilic in nature feeding mainly on deer. Interestingly, An. atroparvus has been implicated in malaria transmission in the UK, however it is highly likely, given its zoophilic preference shown in this study, that its purported anthropophily could have been due to either a shift in the indoor resting place or the high presence of human hosts. Nevertheless, based on previous reports for An. atroparvus and current host data for An. daciae both species could serve as vectors of human borne arboviruses in the UK.

Interestingly, Cx. pipiens, in its current synonymy (chapter 4 Section 4.5), was found to have fed on bird and man. Although this result was obtained from two different individuals in two localities, it shows the potential of Cx. pipiens to feed on both hosts. Culex pipiens is a known vector of WNv in Europe and North America (Chapter 3 section 3.1.4). Higgs et al. 148

(2004) and Medlock et al. (2005) have discussed the potential role that Cx. pipiens and other species could play as vectors in the UK. The identification of hosts, particularly avian and human hosts, from bloodfed Cx. pipiens reported in this thesis (Chapter 4) provides molecular evidence for this hypothesis. Culex torrentium, on the other hand, is reported to be an efficient vector of arboviral disease such as Sindbis (Lundström, 1999), although it is not considered a vector in the UK. Nonetheless its reported and observed (Chapter 5, Figure 5.2 single Cx. torrentium specimen fed on a chaffinch) affinity to birds could indicate a role in the low-level transmission of Sindbis observed in UK chicks (Buckley et al., 2003). In the event of the virus establishing itself in Britain, Cx. torrentium would then serve as a potential vector of the arbovirus.

However, it must be said that hosts identified from bloodfed females collected in this study is not definitive; females could be opportunistic or selective in the choice of hosts based on the prevalence and variety of host species present in a given area (Constantini et al., 1998). This was demonstrated by the diverse feeding selection of An. daciae and the converse for both An. messeae and An. atroparvus specimens sampled in Pettits Animal Farm Norfolk as well as the feeding preference of Cx. pipiens. Thus the ecological data presented here provides a starting point for the incrimination of vectors in the UK, the accurate identification of species and identification hosts including multiple hosts within a bloodmeal. The isolation of a pathogen in the mosquito and host is paramount to the determination and incrimination of vectors.

7.2 Future work

While the data gathered in this study represent a snap shot approach and is by no means exhaustive, this current understanding (prior to an outbreak) provides a much-needed baseline upon which future outbreaks of animal and human diseases could be predicted (Spielman, 1994). Transmission of disease is dependent on several factors: presence of vectors and hosts, climatic conditions and the herd immunity of a population. Based on this and the host selection study presented in this thesis, the introduction of enzoonotic diseases into the UK is tangible. The geographic position of the British Isles, on the fringes of Europe and on the northerly migration route from Africa to Greenland, puts it at risk to the potential introduction of West Nile virus, Sindbis, malaria and Chikungunya via commercial travel (Whitfield et al., 1984; Curtis & White, 1984; Rezza et al., 2007), migratory patterns (Mackenzie et al., 2004), animal movement (Purse et al., 2005; Mellor, 2008) as well as 149 climate change. Global temperatures are expected to increase by at least 6°C by the end of the 21st century (Meteorological Office UK, 2008). This predicted effect of global warming on disease incidence is reported to cause a concomitant increase in humidity and altered rainfall that could be conducive for the development of both vectors and pathogens outside its natural range (Khasnis et al., 2005; Haines et al., 2006).

For example, Aedes albopictus (Stegomyia albopicta) also known as the ‘Asian Tiger mosquito’ is considered to be a highly invasive species in Europe (Gratz, 2004). Indigenous to Southeast Asia (Medlock et al., 2006), it is quickly establishing itself in many parts of Europe such as Spain (Erijta et al., 2005), The Netherlands (Takumi et al., 2008) and Italy (Romi et al., 2001, reviewed in Gratz, 2004). The ability of the eggs to withstand cold temperatures up to -10°C, makes it very adaptive to temperate climates and thus capable of establishing itself in the UK, should it be introduced (Medlock et al., 2006). Also an increase in environmental temperatures can be conducive for the transmission of malaria in the UK. Considering the annual introduction of imported cases of P. falciparum malaria (1,548 cases in 2007) into the UK (DEFRA, 2008), the presence of known vectors An. atroparvus, An. plumbeus, which have been shown to be susceptible to P. falciparum infections and potential (An. daciae) vectors does allow for this possible occurrence.

However, the role of British mosquitoes as vectors should be further established through a multi-factorial approach involving local zoos and farms. As demonstrated in Chapter 5 and by Hutchinson et al. (2007), the Mosquito Magnet Trap ® is a useful tool in monitoring mosquito species in a particular area over a period of time, especially as it was shown to attract species not collected by other means in this study (Oc. detritus & Oc. leucomelas, Chapter 5, section 5.6.2.5). This use of the Mosquito Magnet® in areas such as animal farms [Pettits Animal Farm (Norfolk), Bird World (Isle of Anglesey), Miniature Pony Centre (Devon)] or in domestic farms [such as a Eastlake horse farm (Devon), Cronllech Manor (Anglesey) and Godney Farm (Somerset)], as well as in local zoos [for e.g. Paignton and Jersey Zoos involved in the conservation of non-endemic animals such as Grey Duikers and Madagascan Teal respectively, through the use of studbooks (BIAZA, 2009)] could prove useful in determining interaction of mosquitoes and potential hosts and transmission cycles. These smaller eco-habitats will then facilitate the identification of hosts, using the primers developed in Chapter 4 (Section 4.3.3.2) within the vector blood meal and establish the presence of both viral and parasitic infections in both that host and the vector. This would thus 150

efficiently determine the species of mosquitoes involved in active transmission of arbo- pathogens. It would also tie in with the current effort by the National Expert Panel on New and Emerging Infections (NEPNEI) (Department of Health, 2008) to identify and assess the threat of potential infectious diseases such as malaria, WNv through vector and host surveillance.

All data generated in this study have been submitted to MosquitoMap (www.mosquitomap.org). This freely accessible site stores individual collection records of mosquitoes, complete with georeferenced locality data, method of identification, details of collectors, identifiers, voucher specimen housing and additional fields for detailed ecological and parasite infection data. This database serves to produce risk maps of malaria worldwide using the mosquito species knowledge, environment and human population size using its inbuilt malaria risk assessment software MAL-AREA. This ensures that all data generated in this study are widely accessible and provides a permanent legacy for this research.

Whilst the incidence and threat of introduction of these diseases in the UK are thought to be low, knowledge of local mosquito fauna and vector host interactions should be ascertained and continued surveillance of mosquitoes carried out. In addition, the impact of factors, such as bird migration and climate change on disease transmission in the UK as well as the immunity of the population to mosquito-borne infections, needs to be analysed as interdependent factors. The studies carried out in this thesis contribute towards the current understanding of these ecological processes through the establishment of working laboratory protocols and an ecological database. However, the interpretation of collected ecological results should be done conservatively; as the old medical adage goes “Common things occur commonly, uncommon things don’t. If you hear the sound of hoofbeats think horses, not zebras.”

151

Bibliography

Almeida, A.P.G., Galao, R.P., Sousa, C.A., Novo, M.T., Parreira, R., Pinto, J., Piedade, J. & Esteves, A. (2008) Potential mosquito vectors of arboviruses in Portugal: species, distribution, abundance and West Nile infection. Transactions of the Royal Society of Tropical Medicine and Hygiene, 102, pp. 823-832. Alten, B., Caglar, S.S. & Ozel, O. (2000) Malaria and its vectors in Turkey. European Mosquito Bulletin, 7, pp. 27-33. AMCA Meeting (2002) Discussion forum on the elevation of Aedes and Ochlerotatus. Savannah, Georgia, January 2002. Angelucci, A. (1955) Tavole sinottiche sugli anofelini Italiane. ACIS. Monografia degli Annali di Sanita Pubblica, 1. Apperson, C.S., Harrison, B.A., Unnasch, T.R., Hassan, H.K., Irby, W.S., Savage, H.M., Aspen, S.E., Watson, D.W., Rueda, L.M., Engber, B.R. & Nasci, R.S. (2002) Host- feeding habits of Culex and other mosquitoes (Diptera: Culicidae) in the borough of Queens New York City, with characters and techniques for identification of Culex mosquitoes. Journal of Medical Entomology, 39, pp. 777-785. Arez, A. P. Lopes, D., Pinto, J., Franco, A.S., Snounou, G. & do Rosario, V.E. (2000) Plasmodium sp.: Optimal protocols for PCR detection of low parasite numbers from mosquito (Anopheles sp.) samples. Experimental Parasitology, 94, pp. 269-272. Ashe, P, O'Connor, J.P. & Casey, R.J. (1991) Irish mosquitoes (Diptera, Culicidae): a checklist of the species and their known distribution. Proceedings of the Royal Irish Academy, 91, pp. 21-36. Ashworth, J.H. (1927) The distribution of Anopheline mosquitoes in Scotland. Proceedings of the Royal Society of Edinburgh, 47, pp. 81-93. Atkinson, C.T. (1991) The Encyclopedia of arthropod-transmitted infections. Liverpool School of Tropical Medicine, UK. Bahnck, C. M. & Fonseca, D.M. (2006) Rapid assay to identify the two genetic forms of Culex (Culex) pipiens L. (Diptera: Culicidae) and hybrid populations. American Journal of Tropical Medicine and Hygiene, 75, pp. 251-255. Baldari, M., Tamburro, A., Sabatinelli, G., Romi, R., Severini, C., Cuccagna, G., Fiorilli, G., Allegri, M.P., Buriani, C. & Toti, M. (1998) Malaria in Maremma, Italy. Lancet, 351, pp. 1246-1247.

152

Balenghien, T., Fouque, F., Sabatier, P. & Bicout, D.J. (2006) Horse-, Bird-, and Human- seeking behaviour and seasonal abundance of mosquitoes in a West Nile virus focus of southern France. Journal of Medical Entomology, 43, pp. 936-946. Bardos, V. & Danielova, V. (1959) The Tahyna virus- A virus isolated from mosquitoes in Czechoslovakia. Journal of Hygiene and Epidemiology (Praha), 3, 264-276. Barr, R. (1957) The distribution of Culex p. pipiens and C. p. quinquefasciatus in North America. American Journal of Tropical Medicine and Hygiene, 6, pp. 153-165. Barr, R.A. (1967) Occurrence and distribution of the Culex pipiens complex. Bulletin of the World Health Organization, 37, pp. 293-296. Bates, M. (1939) Variations in the antepalmate hairs of larvae in the Anopheles maculipennis group. Rivista di Malariologia, 18, pp. 299-312. Beach, R., Kilku, G. & Leeuwenburg, J. (1985) Modification of sandfly biting behaviour by Leishmania leads to increased parasite transmission. American Journal of Tropical Medicine and Hygiene, 34, pp. 278-282. Becker, N., Zgomba, M., Petric, D. & Ludwig, M. (1995) Comparison of carbon dioxide, octenol and a host-odour as mosquito attractants in the Upper Rhine Valley, Germany. Medical and veterinary Entomology, 9, pp. 377-380. Beier, J.C., Perkins, P.V., Wirtz, R.A., Koros, J., Diggs, D., Gargan II, T.P. & Koech, D.K. (1988) Blood meal identification by direct Enzyme Linked Immunosorbent Assay (ELISA), tested on Anopheles (Diptera: Culicidae) in Kenya. Journal of Medical Entomology, 25, pp. 9-16. Belkin, J.N. (1962) The mosquitoes of the South Pacific (Diptera: Culicidae). University of California Press. 608pp. Besansky, N.J. & Fahey, T.G. (1997) Utility of the white gene in estimation phylogenetic relationships among mosquitoes (Diptera: Culicidae). Molecular Biology and Evolution, 14, pp. 442-454. Bezzhonova, O.V., & Goryacheva, I.I. (2008) Intragenomic heterogeneity of rDNA internal transcribed spacer 2 in Anopheles messeae (Diptera: Culicidae). Journal of Medical Entomology, 45, pp. 337-341. Blacklock, B. & Carter, H.F. (1920) Further experiments with Anopheles plumbeus, Stephens; its infection with P. falciparum in England; also notes on the apparatus and technique employed. Annals of Tropical Medicine and Parasitology, 14, pp. 275-282.

153

Boakye, D.A., Tang, J., Truc, P., Merriweather, A. & Unnasch, T.R. (1999) Identification of blood meals in haematophagous Diptera by Cytochrome B heteroduplex analysis. Medical and Veterinary Entomology, 13, pp. 282-287. Boccolini, D., Sabatini, A. & Coluzzi, M. (1986) Valore diagnostico del numero dei rami delle setole antepalmate per l’identificazione del complesso Anopheles maculipennis. Annali d’Instituto Superiore di Sanità, 22, pp. 201-204. Boreham, P.F.L. & Garrett-Jones, C. (1973) Prevalence of mixed blood meals and double feeding in a malaria vector (Anopheles sacharovi Favre). Bulletin of the World Health Organization, 48, pp. 605-614. Boreham, P.F.L. (1975) Some applications of blood meal identifications in relation to the epidemiology of vector-borne tropical diseases. Journal of Tropical Medicine and Hygiene, 78, pp. 83-91. Boreham, P.F.L & Lenahan, J.K. (1979) Studies on multiple feeding by Anopheles gambiae s.l. ina Sudan savanna area of north Nigeria. Transactions of the Royal Society of Tropical Medicine and Hygiene, 73, pp. 418- 423. Brown, H.E., Paladini, M., Cook, R.A., Kline, D., Barnard, D. & Fish, D. (2008) Effectiveness of mosquito traps in measuring species abundance and composition. Journal of Medical Entomology, 45, pp. 517-521. Bruce-Chwatt, L.J. & de Zulueta, J. (1980) The rise and fall of malaria in Europe: a historico- epidemiological study. New York. Oxford University Press. pp. 240. Brummer-Korvenkontino, M. (1973) Arboviruses in Finland. V. Serological survey of antibodies against Inkoo virus (California Group) in human, cow, reindeer and wildlife sera. American Journal of Tropical Medicine and Hygiene, 22, pp. 654-661. Bryne, K. & Nichols, R.A. (1999) Culex pipiens in London underground tunnels: differentiation between surface and subterranean populations. Heredity, 82, pp. 7-15. Buckley, A., Dawson, A., Moss, S., Hinsley, S.A., Bellamy, P.A. & Gould, E.A (2003) Serological evidence of West Nile virus, Usutu virus and Sindbis virus infection of birds in the UK. Journal of General Virology, 84, pp. 2807-2817. Buckley A., Dawson, A, & Gould E.A. (2006). Detection of seroconversion to West Nile virus, Usutu virus and Sindbis virus in UK sentinel chickens. Virology Journal, 3, pp. 71.

154

Bullini, L., Bianchi Bullini, A.P., Cianchi, R., Sabatini, A. & Coluzzi, M. (1980) Tassonomia biochima del complessa Anopheles maculipennis. Parassitologia, 22, pp. 290- 292. Bullini, L.A. & Coluzzi, M. (1982) Evolutionary and taxonomic interference of electrophoretic studies in mosquitoes. Recent Developments in the Genetics of Insect Vectors (ed. by Steiner, W.M., Tabachnick, W.J., Rai, K.S. & Narang, S.), Stipes Publishers, Champaign, Illinois. pp. 465-481. Buonomini, G. (1940) L’esame delle setole antepalmate per lo studio della popolazione larvale dei focolai naturali di Anopheles maculipennis. Rivista di Parassitologia, 6, pp. 163- 174. Burkot, T.R. & DeFoliart, G.R. (1982) Blood meal sources of Aedes triseriatus and Aedes vexans in a southern Wisconsin forest endemic for La Crosse Encephalitis virus. American Journal of Tropical Medicine and Hygiene, 31, pp. 376-381. BBC News Online (1999) http://news.bbc.co.uk/2/hi/science/nature/467989.stm (Accessed 12 October 2006) British and Irish Association of Zoos & Aquariums (BIAZA) (2005) http://www.biaza.org.uk/public/pages/projects/index.asp?category=2 (Accessed 12 March 2009). Carn, V.M. (1996) The role of dipterous insects in the mechanical transmission of animal viruses. British Veterinary Journal, 152, pp. 377-393. Carter, H.F. (1978) A list of the Diptera of the Reading area. Reading Naturalist Supplement, 30, pp. 1-70. Chevillon, C., Eritja, R., Pasteur, N & Raymond, M. (1995) Commensalism, adaptation and gene flow: mosquitoes of the Culex pipiens complex in different habitats. Genetic Research Cambridge, 66, pp. 147-157. Cianchi, R., Sabatini, A., Boccolini, D., Bullini, L. & Coluzzi, M. (1987) Electrophoretic evidence of reproductive isolation between sympatric populations of Anopheles melanoon and An. subalpinus. Third International Congress on Malaria and Babesiosis, International Laveran Foundation, 7–11 September, 1987, Annecy, France. Collins, F.H. & Paskewitz, S.M. (1996) A review of the use of ribosomal DNA (rDNA) to differentiate among cryptic Anopheles species. Insect Molecular Biology, 5, pp. 1-9. Corradetti, A. (1934) Richerche sugli incroci tra le varietà di Anopheles maculipennis. Rivista di Malariologia, 13, pp. 707–720.

155

Costantini, C., Sagnon, N., Torre, A.D., Dialo, M., Brady, J., Gibson, G. & Coluzzi, M. (1998) Odor- Mediated host preferences of west African mosquitoes, with particular reference to malaria vectors. American Journal of Tropical Medicine and Hygiene, 58, pp. 56-63. Cranston, P.S., Ramsdale, C.D., Snow, K.R. & White, G.B. (1987) Adults, larvae and pupae of British mosquitoes (Culicidae). Scientific Publications of the Freshwater Biological Association 48. 152pp. Curtis, C.F. & White, G.B. (1984) Plasmodium falciparum transmission in England: entomological and epidemiological data relative to cases in 1983. Journal of Tropical Medicine and Hygiene, 87, pp. 101-114. Cywinska, A., Hunter, F.F. & Herbert, P.D.N (2006) Identifying Canadian mosquitoes species through DNA barcodes. Medical and Veterinary Entomology, 20, pp. 413-424. Da Cunha Ramos, H., Ribiero, H. & Harrison, B.A. (2003) A new European mosquito species: Culex (Neoculex) europaeus (Diptera: Culicidae). European Mosquito Bulletin, 15, pp. 6-11. Dahl, C. (1978) Taxonomic studies on Culex pipiens and C. torrentium in Biosystematics of Haematophagous Insects. Systematics Association Special Volume, 37, pp. 149-175. Davies, C.R. (1990) Interrupted feeding of blood-sucking insects: causes and effects. Parasitology Today, 6, pp 19- 22. de Buck, A., van der Torren, G. & Swellengrebel, N.H. (1933) Report for the year 1932 on investigation into the racial composition of Anopheles maculipennis in Holland. Rivista di Malariologia, 12, pp. 265. de Buck, A. & Swellengrebel, N.H. (1934a) Further observations on the patterns of the upper surface of the ova in the Dutch varieties of A. maculipennis. Proceedings of the Section of Sciences. Koninklijke (Nederlandse) Akademie van Wetenschappen te Amsterdam, 36, pp. 578. de Buck, A. & Swellengrebel, N.H. (1934b) Behaviour of Dutch Anopheles atroparvus and messeae in winter under artificial conditions. Rivista di Malariologia, 8, pp. 404. de Zulueta, J., Ramsdale, C.D., Cianchi, R., Bullini, L. & Coluzzi, M. (1983) Observations on the taxonomic status of Anopheles sicaulti. Parassitologia, 23, pp. 73-92. Defoliart, G.R., Grinstad, P.R. & Watts, D.M. (1987) Advances in mosquito-borne arbovirus/vector research. Annual Review in Entomology, 32, pp. 479-505. DeSalle, R., Egan, M.G. & Siddall, M. (2005) The unholy trinity: taxonomy, species delimitation and DNA barcoding. Philosophical Transactions of The Royal Society series B, 360, pp. 1905-1916. 156

Del Giudice, P., Schuffenecker, I., Vandenbos, F., Counillon, E. &, Zeller, H. (2004) Human West Nile Virus, France. Emerging Infectious Diseases, 10, pp. 1885-1886. Deruaz, D., Deruaz, J. & Pichot, J. (1991) Correspondence analysis of larval chaetotaxy in the "Anopheles maculipennis complex" (Diptera: Culicidae). Annales de Parasitologie Humaine et Comparée, 66, pp. 166-172. Detinova, T.S. (1953) Age composition and epidemiological importance of the population of Anopheles maculipennis in the Province of Moscow. Meditsinskaya Parazitologiya i Parazitarnie Bolezni, 22, pp. 486-495. Di Luca, M., Boccolini, D., Marinucci, M. & Romi, R. (2004) Intrapopulation polymorphism in Anopheles messeae (An. maculipennis complex) inferred by molecular analysis. Journal of Medical Entomology, 41, pp. 582-586. Diaz, S., Novo, M.T., Sousa, C.A., Alves, R., Almeida, A.P.G. & Pinto, J. (2006) A biological study of the Culex pipiens complex in Comporta, Portugal. Abstract, 15th European Society of Vector Ecology Meeting, 10-14 April 2006, Serres, Greece. Diemer, J. (1935) La méthode de Heincke de combinaison des caractères pour la détermination raciale d’un exemplaire isolé d’Anopheles maculipennis. Bulletin de la Societe de Pathologie Exotique, 28, pp. 932–936. Djadid, N.D., Gholizadeh, S., Tafsiri, E., Romi, R., Gordeev, M. & Zakeri, S. (2007) Molecular identification of Palaearctic members of Anopheles maculipennis in northern Iran. Malaria Journal, 6, pp. 1-10. Dobson, M.J. (1989) History of malaria in England. Journal of the Royal Society of Medicine, 82 (S17), pp. 3-7. Dutta, P., Khan, S.A., Sharma, C.K. & Mahanta, J. (1997) A report of mosquito fauna survey and vector incrimination in Goalpara district of Assam. Indian Journal of Malariology, 34, pp. 204-207. Department for Environment, Food and Rural Affairs (DEFRA) (2008) http://www.defra.gov.uk/animalh/diseases/vetsurveillance/az_index.htm (Accessed 30 January 2009) Department of Health (2004) http://www.dh.gov.uk/en/Aboutus/MinistersandDepartmentLeaders/ChiefMedicalOffic er/ProgressOnPolicy/ProgressBrowsableDocument/DH_5014877 (Accessed 13 June 2008). Edwards, F.W. (1932) Genera insectorum. Diptera. Family Culicidae. Fascicle 194. Belgium. Pp. 258. 157

Edwards, F.W. (1936) Probable occurrence in England of the so-called typical race of Anopheles maculipennis. Entomologist, 64, pp. 242-243. Eling, W., van Gemert, G.-J., Akinpelu, O., Curtis, J. & Curtis, C. F. (2003) Production of Plasmodium falciparum sporozoites by Anopheles plumbeus. European Mosquito Bulletin, 15, pp. 12-13. Eritja, R., Escosa, R., Lucientes, J., Marques, E., Molina, R., Roiz, D. & Ruiz, S. (2005) Worldwide invasion of vector mosquitoes: present European distribution and challenges for Spain. Biological Invasions, 7, pp. 87-97. Esteves, A., Almeida, P.G., Galao, R.P., Parreira, R., Peidade, J., Rodrigues, J.C., Sousa, C.A. & Novo, T.M. (2005) West Nile in Southern Portugal, 2004. Vector-Borne and Zoonotic Diseases, 5, pp. 410-413. Evans, A.M. (1934) On the differentiation of Anopheles maculipennis in Great Britain, with special reference to a form occurring on the coast of North Wales. Annals of Tropical Medicine and Parasitology, 28, pp. 131-140. Falleroni, D. (1926) Fauna anofelica italiana e suo 'habitat' (paludi, risaie, canali). Metodi di lotta contro la malaria. Rivista di Malariologia, 5, pp. 553-59. Falleroni, D. (1932) Zooprofilassi e sua applicazione in Ardea. (Argo Romano). Diverse razze di "A. maculipennis". Tipografia Cuggiani, Roma. pp. 48+ x. Fantini, B. (1994) Anophelism without Malaria: an ecological and epidemiological puzzle. Parassitologia, 36, pp. 83-106. Fernandez, R.J. & Vazquez, S. (1990) Serological diagnosis of dengue by an ELISA inhibition method (EIM). Memorias do Instituti Oswaldo Cruz, 85, pp. 347-351. Field, L.M., Devonshire, A.L., French-Constant, R.H. & Forde, B.G. (1989) Changes in DNA methylation are associated with loss of insecticide resistance in the peach-potato aphid Myzus persicae. FEBS letters, 243, pp. 323-327. Field, L.M., Lyko, F., Mandrioli, M. & Prantera, G. (2004) DNA methylation in insects. Insect Molecular Biology, 13, pp. 109-115. Fonseca, D.M., Keyghobadi, N., Malcolm, C.A., Mehmet, C., Schaffner, F., Mogi, M., Fleischer, R.C. and Wilkerson, R.C. (2004) Emerging vectors in the Culex pipiens complex. Science, 303, pp. 1535-1538. Folmer, O., Black, M., Hoeh, W., Lutz, R. & Vrijenhoek, R. (1994) DNA primers for amplification of mitochondrial Cytochrome c Oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology, 3, pp 294-299.

158

Francy, D. B., Jaeson, T.G.G., Lundström, J.O., Espmark, Å., Schildt, E. B., Henriksson, B. & Niklasson, B. (1989) Ecologic studies of mosquitoes and bird as hosts of Ockelbo virus in Sweden and isolation of Inkoo and Batai viruses from mosquitoes. American Journal of Tropical Medicine and Hygiene, 41, 355-363. Freed, L.A & Cann, R.L. (2003) On polymerase chain reaction tests for estimating prevalence of malaria in birds. Journal of Parasitology, 89, pp. 1261-1264. Fritz, G.N., Conn, J., Cockburn, A. & Seawright, J. (1994) Sequence analysis of the ribosomal DNA internal transcribed spacer 2 from a population of Anopheles nuneztovari (Diptera: Culicidae). Molecular Biology and Evolution, 11, pp. 406-416. Frizzi, G. (1952) Nuovi contributi e prospetti di ricerca nel gruppo Anopheles maculipennis in base allo studio del dimorfismo cromosomico. Symposia Genetica, 3, pp. 231-265. Frizzi, G. (1953) Etude cytogénétique d'Anopheles maculipennis en Italie. Bulletin of the World Health Organization, 9, pp. 335-344. Fyodorova, M.V., Savage, H.M., Lopatina, J.V., Bulgakova, T.A., Ivanitsky, A.V., Platonova, O.V. & Platonov, A.E. (2006) Evaluation of potential West Nile virus vectors in Volgograd Region, Russia, 2003 (Diptera: Culicidae): species composition, blood meal host utilisation and virus infection rates of mosquitoes. Journal of Medical Entomology, 43, pp. 552-563. Gillies, M.T. & Gubbins, S.J. (1982) Culex (Culex) torrentium Martini and Cx. (Cx.) pipiens L. in the southern English county, 1974-1975. Mosquito Systematics, 14, pp. 127-130. Goloboff, P.A., Farris, J.S. & Nixon, K.C. (2008) TNT, a free program for phylogenetics analysis. Cladistics, 24, pp. 774-786. Gordeev, M., Goriacheva, I., Shaikevich, E. & Ejov, M. (2004) Variability of the second internal transcribed spacer of the ribosomal DNA among five Palaearctic species of Anopheline mosquitoes. European Mosquito Bulletin, 17, pp. 14-19. Gordeev, M.I., Zvantsov, A.B., Goriacheva, I.I., Shaikevich, E.V. & Ezhov, M.N. (2005) Description of the new species Anopheles artemievi sp. n. (Diptera, Culicidae). Meditsinskaia parazitologiia i parazitarnye bolezni, 2, pp. 4-5. Gordon, S.W., Tammariello, R.F., Linthicum, K.J., Wirtz, R.A. & Digoute, J.P. (1991) Feeding patterns of mosquitoes collected in the Senegal river basin. Journal of the Americal Mosquito Control Association, 7, pp. 424- 432. Gould, E.A, Higgs, S., Buckley, A. & Gritsun, T.S. (2006) Potential arbovirus emergence and implications for the United Kingdom. Emerging Infectious Diseases, 12, pp. 549-555.

159

Gratz, N.G. (2004) Critical review of the vector status of Aedes albopictus. Medical and Veterinary Entomology, 18, pp. 215-227. Guidon, S. & Gascuel, O. (2003) A simple, fast and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology, 52, pp. 696-704. Gutsevich, A.V., Monchadskii, A.S. & Shtakel’berg, A.A. (1974) Fauna of the USSR Diptera, Vol. 3 No. 4 Mosquitoes Family Culicidae, Keter Publishing House Jerusalem Ltd. Jerusalem pp. 480. Guy, Y., Salières, A & Boesiger, E. (1976a) Contribution à l’étude du “complexe maculipennis” (Diptera–Culicidae–Anophelinae) mise au point en 1975. Année Biologique, 15, pp. 227-282. Guy, Y., Salieres, A. & Lumaret, R. (1976b) Polymorphisme de I’ornementation des oeufs chez Anopheles melanoon subalpinus (complexe maculipennis Diptera-Culicidae- Anophelinae). Compte Rendu de L’Academie des Sciences Paris, 285, pp. 809-811. Hackett, L.W. & Lewis, D.J. (1935) A new variety of Anopheles maculipennis in southern Europe. Rivista di Malariologia, 14, pp. 377-383. Hackett, L.W. & Missiroli, A. (1935) The varieties of Anopheles maculipennis and their relation to the distribution of malaria in Europe. Rivista di Malariologia, 14, pp. 45- 109. Hadfield, T.L., Turrell, M., Dempsey, M.P., David, J. & Park, E.J. (2001) Detection of West Nile virus in mosquitoes by RT-PCR. Molecular and Cellular Probes, 15, pp.147-150. Haines, A., Kovats, R.S., Campbell-Lendrum, D. & Corvalan, C. (2006) Climate change and human health: impacts, vulnerability and public health. Public Health, 120, pp. 585- 596. Hamer, G.L., Kitron, U.D., Brawn, J.D., Loss, S.R., Ruiz, M.O., Goldberg, T.L. & Walker, E.D. (2008) Culex pipiens (Diptera: Culicidae): A bridge vector of West Nile virus to humans. Journal of Medical Entomology, 45, pp. 125-128. Hannoun, C., Panthier, R., Mouchet, J. & Eouzan, J-P. (1964) Isolement en France du virus West-Nile a’patir de maladies et du vecteur Culex modestus Ficalbi. Comptes rendus des séances de l’Academie des sciences. Serie D. Sciences naturelles, 259, pp. 4170- 4172. Harbach, R.E. (1988) The mosquitoes of subgenus Culex in southwestern Asia and Egypt (Diptera: Culicidae). Contributions of the American Entomological Institute, 24, pp. 1- 240.

160

Harbach, R.E. (2004) The classification of genus Anopheles (Diptera: Culicidae): a working hypothesis of phylogenetic relationships. Entomological Research, 94, pp. 537-553. Harbach, R.E. & Knight, K.L. (1980) Taxonomists' Glossary of Mosquito Anatomy. Plexus Publishing, Inc., Marlton, New Jersey. 415 pp. Harbach, R.E., Harrison, B.A. & Gad, A.M. (1984) Culex (Culex) molestus Forksål (Diptera: Culicidae): Neotype designation, description, variation and taxonomic status. Proceedings of the Entomological Society of Washington, 86, pp. 521-542. Harbach, R.E., Dahl, C. & White, G.B. (1985) Culex (Culex) pipiens Linnaeus (Diptera: Culicidae): concepts, type, designations and description. Proceedings of the Entomological Society of Washington, 87, pp. 1-24. Harbach, R.E. & Kitching, I.J. (1998) Phylogeny and classification of the Culicidae (Diptera). Systematic Entomology, 23, pp. 327-370. Harbach, R.E. & Kitching, I.J. (2005) Reconsideration of Anopheline mosquito phylogeny (Diptera: Culicidae: Anophelinae) based on morphological data. Systematics and Biodiversity, 3, pp. 435-374. Harbach, R.E. & Howard, T.M. (2007) Index of currently recognized mosquito species (Diptera: Culicidae). European Mosquito Bulletin, 23, pp. 1-66. Hart, G.F.W (1954) Rediscovery of Anopheles algeriensis Theobald (Diptera: Culicidae) in Norfolk. Entomologist’s Monthly Magazine, 90, 63. Hayes, C.G. (2001) West Nile Virus: Uganda, 1937, to New York City, 1999. Annals of the New York Academy of Sciences, 951, pp. 25-37. Hebert, P.D.N., Penton, E.H., Burns, J.M., Janzen, D. & Hallwachs, W. (2004) Ten species in one: DNA barcoding reveals cryptic species in neotropical skipper butterfly Astraptes fulgerator. Proceedings of the National Academy of Sciences, 12, pp. 14812-14817. Heid, C.A., Stevens, J. & Livak, K.J & Williams, P.M. (1996) Real time quantitative PCR. Genome Research, 6, pp. 986-994. Hick, C.A., Field, L.M. and Devonshire, A.L. (1996) Changes in the methylation of amplified esterase DNA during loss and reselection of insecticide resistance in peach-potato aphids, Myzus persicae. Insect Biochemistry and Molecular Biology, 26, pp. 41-47. Higgs, S., Snow, K. & Gould, E.A. (2004) The potential for West Nile virus to establish outside of its natural range: a consideration of potential mosquito vectors in the United Kingdom. Transactions of the Royal Society of Tropical Medicine and Hygiene, 98, pp. 82-87.

161

Hubalek, Z. & Halouzka, J. (1996) Arthropod-Borne viruses of vertebrates in Europe. Acta Scientiarum Naturalium Academiae Scientiarum Bohemicae-Brno, 4-5, pp. 95 Hubalek, K. Z. & Halouzka J. (1999) West Nile fever a re-emerging mosquito-borne viral disease in Europe. Emerging Infectious Diseases, 5, pp. 643-650. Huff, C.G. (1965) Susceptibility of mosquitoes to avian malaria. Experimental Parasitology, 16, pp. 107-132. Hutchinson, R.A., West, P.A. & Lindsay, S.W. (2007) Suitability of two carbon dioxide-baited traps for mosquito surveillance in the United Kingdom. Bulletin of Entomological Research, 97, pp. 591-597. Health Protection Agency (HPA) (2008) http://www.hpa.org.uk/webw/HPAweb&Page&HPAwebAutoListName/Page/1191942 128239?p=1191942128239 (Last accessed 27 February 2009). Işfan, Tr.N. (1952) Chetotaxia nimfelor in determinarea raselor şi hibrizilor de Anopheles maculipennis. Buletin Ştiinţific, 4, pp. 717-732. Ishtiaq, F., Gerin, E., Rappole, J.H., Rahmani, A.R., Jhala, Y.V., Dove, C.J., Milensky, C., Olson, S.L., Peirce, M.A. & Fleischer, R.C. (1007) Prevalence and diversity of avian hematozoan parasites in asia: a regional survey. Journal of Wildlife Diseases, 43, pp. 382-398. Jaenson, T.G.T., Lokki, J. & Saura, A. (1986a) Anopheles (Diptera: Culicidae) and malaria in northern Europe, with special reference to Sweden. Journal of Medical Entomology 23: 68-75. Jaenson, T.G.T & Niklasson, B. (1986b) Feeding patterns of mosquitoes (Diptera:Culicidae) in relation to the transmission of Ockelbo disease in Sweden. Bulletin of Entomological Research, 76, pp. 375-383. James, S.P. (1929) The disappearance of malaria from England. Proceedings of the Royal Society for Medicine, 23, pp 71-85. Janini, R., Saliba, E., Khoury, S., Oumeish, O., Adwan, S & Kamhawi, S. (1995) Incrimination of Phlebotomus papatasi as vector of Leishmania major in the southern Jordan Valley. Medical and Veterinary Entomology, 9, pp. 420-422. Jetten, T.H. & Takken, W. (1994) Anophelism without malaria in Europe. A review of the ecology and distribution of the genus Anopheles in Europe. Wageningen Agricultural University Papers, p. 94-5. Jetten, T.H., Martens, W.J.M. & Takken, T. (1996) Model simulations to estimate malari risk under climate change. Journal of Medical Entomology, 33, pp. 61-371. 162

Johansen, C.A., Hall, R.A., Van den Hurk, A.F., Ritchie, S.A. & Mackenzie, J.S. (2002) Detection and stability of Japanese Encephalitis virus RNA and virus viability in dead infected mosquitoes under different storage conditions. American Journal of Tropical Medicine and Hygiene, 67, pp. 656-661. Jupp, P.G. (1979) Culex (Culex) pipiens and Culex (Cx.) torrentium (Diptera: Culicidae) in England: notes on their taxonomy and biology. Mosquito Systematics, 11, pp. 121-126. Kampen, H. (2005a) The ITS2 ribosomal DNA of Anopheles beklemishevi and further remarks on the phylogenetic relationships within the Anopheles maculipennis group of species (Diptera: Culicidae). Parasitology Research, 97, pp. 118-128. Kampen, H. (2005b) Integration of Anopheles beklemishevi (Diptera: Culicidae) in a PCR assay diagnostic for Palaearctic Anopheles maculipennis sibling species. Parasitology Research, 97, pp. 113-117. Kasap, M. (1990) Comparison of experimental infectivity and development of Plasmodium vivax in Anopheles sacharovi and An. superpictus in Turkey. American Journal of Tropical Medicine and Hygiene, 42, pp. 111-117. Kent, R.J. & Norris, D.E. (2005) Identification of mammalian blood meals in mosquitoes by a multiplexed polymerase chain reaction targeting Cytochrome B. American Journal of Tropical Medicine and Hygiene, 73, pp. 336-342. Khasnis, A.A. & Nettleman, M.D (2005) Global warming and infectious disease. Archives of Medical Research, 36, pp. 689-696. Killington, F.J. (1946) Mosquitoes of the Poole district. Journal of the Society of British Entomology, 3, pp. 35-36. Kilpatrick, A.M., Kramer, L.D., Jones, M.J, Marra, P.P., Daszak, P. & Fonseca, D.M. (2007) Genetic influences on mosquito feeding behaviour and the emergence of zoonotic pathogens. American Journal of Tropical Medicine and Hygiene, 77, pp. 667-671. Kirstein, F. & Gray, J.S. (1996) A molecular marker for the identification of the zoonotic reservoirs of Lyme Borreliosis by analysis of the blood meal in its European vector Ixodes ricinus. Applied and Environmental Microbiology, 62, pp. 4060-4065 Kitzmiller, J.B., Frizzi, G. & Baker, R. (1967) Evolution and speciation within the maculipennis complex of the genus Anopheles. Genetics of Insect Vectors of Disease (ed. by J.W. Wright & R. Pal), Elsevier Publishing, London. pp. 151-210. Knight, K.L. & Stone, A. (1977) A catalog of the mosquitoes of the world. 2nd edition Thomas say Foundation 611pp.

163

Knols, B.G.J., Takken, W., Cork, A. & De Jong, R. (1997) Odour-mediated, host-seeking behaviour of Anopheles mosquitoes: a new approach. Annals of Tropical Medicine and Parasitology, 91, S117-S118. Korvenkontio, P., Lokki, J., Saura, A. & Ulmanen, I. (1979) Anopheles maculipennis complex (Diptera: Culicidae) in northern Europe: species diagnosis by egg structure and enzyme polymorphism. Journal of Medical Entomology, 16, pp. 169-170. Krishnamurthy, B.S. (1961) A note on inheritance of autogeny in Culex mosquitoes. Bulletin of the World Health Organisation, 24, pp. 675-677. Krzywinski, J., Wilkerson, R.C. & Besansky, N.J. (2001a) Evolution of mitochondrial and ribosomal gene sequences in Anophelinae (Diptera: Culicidae): implications for phylogeny reconstruction. Molecular Phylogenetics and Evolution, 18, pp. 479-487. Krzywinski, J., Wilkerson, R.C. & Besansky, N.J. (2001b) Toward understanding Anophelinae (Diptera, Culicidae) phylogeny: insights from nuclear single-copy genes and the weight of evidence. Systematic Biology, 50, pp. 540-556. Kuhn, K.G., Campbell-Lendrum, D.H., Armstrong, B. & Davies, C.R. (2003) Malaria in Britain: Past, present and future. Proceedings of the National Academy of Sciences, 100, 9997. Photo source: Copyright (2008) National Academy of Sciences, U.S.A. Kumar, N.P., Rajavel, A.R., Natarajan, R. & Jambulingam, P. (2007) DNA Barcodes Can Distinguish Species of Indian Mosquitoes (Diptera: Culicidae). Journal of Medical Entomology, 44, pp. 1-7. Lvov, D.K., Gromashevsky, V.L., Sidora, G.A., Tsyrkin, Y.M., Chervonsky, V.I. & Asritova, V.A. (1972) Isolation of Tahyna virus from Anopheles hyrcanus mosquitoes in Kyzylagach preserve, south-easter Azerbaijan. Voprosy virusologii, 1, 18-21. L’vov, S.D. Gromashevskii, V.L. Skvortsova, N.G., Andranova, O.N., Semenov, V.B., Gushchina, E.A., Morozova, T.N., Kuznetsov, A.A., Glakina, I.V., Yamnikova, S.S., Aristova, V.A., Avershin, A.D. & Berezin, M.V. (1989) Circulation of viruses from the Californian encephalitis complex in the northwest of the Russian Plain. Meditsinskaya Parazitologiya i Parazitamye Bolezni, 2, pp. 74-77. Lvov, D.K., Butenko, A.M., Gromashevsky, V.L., Larichev, V.Ph., Gaidamovich, S.Ya., Vyshemirsky, O.I., Zhukov, A.N., Lazorenko, V.V., Salko, V.N., Kovtunov, A.I., Galimzyanov, Kh.M., Platonov, A.E., Morozova, T.N., Khutoretskaya, N.V., Shishkina, E.O. & Skvortsova, T.M. (2000) Isolation of two strains of West Nile virus during an outbreak in southern Russia, 1999. Emerging Infectious Diseases, 6, pp. 373- 376. 164

La Face, L. (1931) Sull’esistenza di razze diverse di Anopheles maculipennis. Rivista di Malariologia, 18, pp. 299–312. Lang, W.D. (1918) A map showing the known distribution in England and Wales of the Anopheline mosquitoes, with explanatory text and notes. British Museum (Natural History), London. 63pp + map. Lang, W.D. (1920) A handbook of British mosquitoes. British Museum (Natural History), London. 125pp + 4pl. Lanicotti, R.S., Roehrig, J.T., Deubel, V., Smith, J., Parker, M., Steele, K., Crise, B., Volpe, K.E., Crabtree, M.B., Scherret, J.H., Hall, R.A., Mackenzie, J.S., Cropp, C.B., Panigrahy, B., Ostlund, E., Schmitt, B., Malkinson, M., Banet, C., Weissman, J., Komar, N., Savage, H.M., Stone, W., McNamara, T. & Gubler, D.J. (1999) Origin of the West Nile virus responsible for an outbreak of encephalitis in the northeastern United States. Science, 286, pp. 2333-2337. Lee, A., Hutchinson, R. & Nikolaeva, N.V. (2002) A comparison of Anopheles messeae in the UK and Russia. Transactions of the Royal Society for Tropical Medicine and Hygiene, 96, pp. 362. Lee, J.H., Hassan, H., Hill, G., Cupp, E.W., Higazi, T.B., Mitchell, C.J., Godsey Jr., M.S. & Unnasch, T.R. (2002) Identification of mosquito avian-derived blood meals by Polymerase Chain Reaction - heteroduplex analysis. American Journal of Tropical Medicine and Hygiene, 66, pp. 299-604. Lindsay, S.W. & Birley, M.H. (1996) Climate change and malaria transmission. Annals of Tropical Medicine and Parasitology , 90, pp. 573-588. Linton, Y.-M, Samanidou-Voyadjoglou, A., Smith, L. & Harbach R.E. (2001a) New occurrence records for Anopheles maculipennis and An. messeae in northern Greece based on DNA sequence data. European Mosquito Bulletin, 11, pp. 31-36. Linton, Y.-M., Harbach, R.E., Chang, M.S., Anthony, T.G. & Matusop, A. (2001b) Morphological and molecular identity of Anopheles (Cellia) sundaicus (Diptera: Culicidae), the nominotypical member of a malaria vector species complex in Southeast Asia. Systematic Entomology, 26, pp. 357-366. Linton, Y.-M., Smith, L. & Harbach, R.E. (2002a) Molecular confirmation of sympatric populations of Anopheles messeae and Anopheles atroparvus overwintering in Kent, southeast England. European Mosquito Bulletin, 13, pp. 8-16.

165

Linton, Y.-M., Smith, L. & Harbach, R.E. (2002b) Observations on the taxonomic status of Anopheles subalpinus Hackett & Lewis and An. melanoon Hackett. European Mosquito Bulletin, 13, pp. 1-7. Linton, Y.-M., Samanidou-Voyadjoglou, A. & Harbach, R.E. (2002c) Ribosomal ITS2 sequence data for Anopheles maculipennis and An. messeae in northern Greece, with a critical assessment of previously published sequences. Insect Molecular Biology, 11, pp. 379-383. Linton, Y.-M., Smith, L., Koliopoulos, G., Samanidou-Voyadjoglou, A., Zounos, A.K. & Harbach R.E. (2003) Morphological and molecular characterization of Anopheles (Anopheles) maculipennis Meigen, type species of the genus and nominotypical member of the Maculipennis Complex. Systematic Entomology, 28, pp. 36-56. Linton, Y.-M. (2004) Systematics of the Holarctic Maculipennis Complex. Abstract booklet of the 70th Annual Meeting of the American Mosquito Control Association, 22–26 February, 2004, Savannah, Georgia, USA. Linton, Y.-M., Lee, A.S. & Curtis, C. (2005) Discovery of a third member of the Maculipennis Group in SW England. European Mosquito Bulletin, 19, pp. 5-8. Linton, Y.-M., Smith, L., Koliopoulos, G., Zounos, A.K., Samanidou-Voyadjoglou, A., Patsoula, E. & Harbach, R.E. (2007) The Anopheles maculipennis complex (Diptera: Culicidae) in Greece. Journal of Natural History , 41 Lundström, J.O., Niklasson, B. & Francy, B.D. (1990a) Swedish Culex torrentium and Cx. pipiens (Diptera: Culicidae) as experimental vectors of Ockelbo virus. Journal of Medical Entomology, 27, pp. 561-563. Lundström, J.O., Turell, M.J. & Niklasson, B. (1990b) Effect of environmental temperature on the vector competence of Culex pipiens and Cx. torrentium for Ockelbo virus. American Journal of Tropical Medicine and Hygiene, 43, pp. 534-542. Lundström, J.O. (1999) Mosquito-Borne viruses in western Europe: a review. Journal of Vector Ecology, 24, pp. 1-39. Lundström, J.O., Lindstrom, K.M., Olsen, B., Dufva, R. & Krakower, D.S. (2001) Prevalence of Sindbis virus neutralizing antibodies among Swedish Passerines indicates that Thrushes are the main amplifying hosts. Journal of Medical Entomology, 38, pp. 289- 297. Mackenzie, J.S., Gubler, D.J. & Petersen, L.R. (2004) Emerging flaviviruses: the spread and resurgence of Japanese encephalitis West Nile and dengue viruses. Nature Medicine Supplement, 10, S98-S109. 166

Mandrioli, M. (2004) Epigenetic tinkering and evolution: is there any continuity in the role of cytosine methylation from invertebrates to vertebrates? Cellular and Molecular Life Science, 61, pp. 2425-2427. Mangold, K.A., Manson, R.U, Koay, E.S.C., Stephens, L., Regner, M., Thomson Jr., R.B., Peterson, L.R. & Kaul, K.L. (2005) Real-time PCR for detection and identification of Plasmodium spp. Journal of Clinical Microbiology, 43, pp. 2435-2440. Marchant, P. Eling, W., van Gemert, G.-J., Leake, C.J. & Curtis, C.F. (1998) Could British mosquitoes transmit falciparum malaria? Parasitology Today, 14, pp. 344- 345. Marfin, A.A., Petersen, L.R., Eidson, M., Miller, J., Hadler, J., Farello, C., Werner, B., Campbell, G.L., Layton, M., Smith, P., Bresnitz, E., Cartter, M., Scaletta, J., Obiri, G., Bunning, M., Craven, R.C., Roehrig, J.T., Julian, K.G., Hinten, S.R., Gubler, D.J. & ArboNET Cooperative Surveillance Group (2001) Widespread West Nile virus activity, eastern United States, 2000. Emerging Infectious Diseases, 7, pp. 730-735 Marinucci, M., Romi, R., Mancini, P., Di Luca, M. & Severini, C. (1999) Phylogenetic relationships of seven Palaearctic members of the maculipennis complex inferred from ITS2 sequence data. Insect Molecular Biology, 8, pp. 469-480. Martini, E. 1925 Zwei bermerkenswerte Culiciden von einem eigenartigen Biotop Internationale Revue der Gesamten Hydrobiologie und Hydrographie, 12, pp. 333- 337. Marshall, J.F. & Staley, J. (1933) Variations in the surface pattern of eggs of Anopheles maculipennis (Diptera: Culicidae) obtained in the south of England, Stylops, 2, pp. 238-240. Marshall, J.F. & Staley, J. (1937). Some notes regarding the morphological and biological differentiation of Culex pipiens Linnaeus and Culex molestus Forksål (Diptera, Culicidae). Proceedings of the Royal Entomological Society of London 12: 17-26. Marshall, J.F. (1938) The British mosquitoes London. British Museum (Natural History). xi +341 pp. Marshall, J.F. (1945) Records of Culex (Barraudius) modestus obtained in the South of England. Nature London, 156, pp. 172-173. Mattingly, P.F (1951) Culex (Culex) torrentium Martini, a mosquito new to Great Britain. Nature, 168, pp. 172-173. Mattingly, P.F. (1950) Family Culicidae, subfamily Culicine. In “Diptera 2: Nemocerca”. Handbook for the identification of British Insects, 9, pp. 102-120.

167

Mattingly, P.F. (1952) The problem of biological races in the Culex pipiens complex. Proceedings of the Linnaean Society of London, 163, pp. 53-55. Mattingly, P.F. (1953) A change of name among the British mosquitoes. Proceedings of the Royal Society of London Series B, 22, pp. 106-108. Mattingly, P.F (1967) The systematics of the Culex pipiens complex. Bulletin of the World Health Organization, 37, pp. 257-261. Medlock, J.M., Avenell, D., Barrass, I. & Leach, S. (2006) Analysis of the potential for survival and seasonal activity of Aedes albopictus (Diptera: Culicidae) in the United Kingdom. Journal of Vector Ecology, 31, pp.292-304. Medlock, J.M., Snow, K.R. & Leach, S. (2005) Potential transmission of West Nile virus in the British Isles: an ecological review of candidate mosquito bridge vectors. Medical and Veterinary Entomology, 19, pp. 2-21. Medlock, J.M., Snow, K.R. & Leach, S. (2007) Possible ecology and epidemiology of medically important mosquito-borne arboviruses in Great Britain. Epidemiology and Infection, 135, pp. 466-482. Medlock, J.M. & Vaux, A.G.C. (2009) Aedes (Aedes) geminus Peus (Diptera, Culicidae) – an addition to the British mosquito fauna. Dipsterists Digest, 16, pp. 147-150. Meece, J.K., Reynolds, C.E., Stockwell, P.J., Jenson, T.A., Christensen, J.E. & Reed K. D. (2005) Identification of mosquito blood meal source by terminal restriction fragment length polymorphism profile analysis of the Cytochrome B gene. Journal of Medical Entomology, 42, pp 657-667. Meeghan, J.M., Khalil, G.M., Hoogstraal, H. & Adham, F.K. (1980) Experimental transmission and field isolation studies implicating Culex pipiens as a vector of Rift Valley fever virus in Egypt. American Journal of Tropical Medicine and Hygiene, 29, pp. 1405-1410. Mellor, P.S. (2008) Bluetongue in Europe: vectors, epidemiology and climate change. 16th European Society for Vector Ecology Conference 2008, 25-28 March 2008, Cambridge. Meyer, R.P., Hardy, J.L. &Presser, S.B. (1982) Variation in vector competence of Culex pipiens to St. Louis Encephalitis virus. Proceedings Paper from the 50th Annual Conference of Californian Mosquito Vector Control Association. Michael, E., Ramaiah, K.D., Hoti, S.L., Barker, G., Paul, M.R., Yuvaraj, J., Sa, P.K., Grenfell, B.T. & Bundy, A. P. (2001) Quantifying mosquito biting patterns on humans by DNA

168

finerprinting of bloodmeals. American Journal of Tropical Medicine and Hygiene, 65, 722-728. Miller, B.R., Crabtree, M.B. & Savage, H.M. (1996). Phylogeny of fourteen Culex mosquito species, including the Culex pipiens complex, inferred from the internal transcribed spacers of ribosomal DNA. Insect Molecular Biology, 5, pp. 93-107. Mohrig, W (1969) Die Culiciden Deutschlands. Untersuchungen zur taxonomie, Biologie und Okologie der einheimischen steckmucken. Parasitologische Schriftenreihe, 260, pp. 18. Morgan, M. J. (1978) Mosquitoes in North Wales. Nature in Wales, 16, pp. 102-104. Morgan, M.J. (1987) Re-appearance of Anopheles algeriensis Theobald in Britain. British Medical Group Newsletter, 4, pp. 1. Muirhead-Thomson, R.C. (1956) Field studies of the role of Anopheles atroparvus in the transmission of myxomatosis in England. Journal of Hygiene, 54, pp. 472-477. Mukherjee, N., Mondol, S., Andheria, A. & Ramakrishnan, U. (2007) Rapid multiplex PCR based species identification of wild tigers using non-invasive sampling. Conservation Genetics, 8, pp. 1465-1470. Mullen, G. & Durden, L. (2002) Medical and Veterinary Entomology. Elsevier Science, USA 597pp. Munstermann, L.E. & Conn, J.E. (1997) Systematics of mosquito disease vectors (Diptera, Culicidae): impact of molecular biology and cladistic analysis. Annual Reviews in Entomology, 42, pp. 351-369. Meterological Office UK (2008) http://www.metoffice.gov.uk/corporate/pressoffice/2008/pr20080326.html (Accessed 31 March 2009). Mosquito Taxonomic Inventory (2008) http://mosquito-taxonomic- inventory.info/taxonomy/term/6045 (Accessed 22 October 2008). Navajas, M., Lagnel, J., Gutierrez, J. & Boursot§, P. (1998) Species-wide homogeneity of nuclear ribosomal ITS2 sequences in the spider mite Tetranychus urticae contrasts with extensive mitochondrial COI polymorphism. Heredity, 80, pp. 742-752. Navarro, J.-C. & Liria, J. (2000) Phylogenetic relationships among eighteen Neotropical Culicini species. Journal of the American Mosquito Control Association, 16, pp. 75-85. Nicolescu, G., Linton, Y.-M., Vladimirescu, A.F., Howard, T.M. & Harbach, R.E. (2004) Mosquitoes of the Anopheles maculipennis group (Diptera: Culicidae) in Romania,

169

with the discovery and formal recognition of a new species based on molecular and morphological evidence. Bulletin of Entomological Research, 94, pp. 525-535. Niklasson, B., Espmakr, Å., LeDuc, J.W., Gargan, T.P., Ennis, W., Tesh, R.B. & Main Jr., A.J. (1984) Association of Sindbis-like virus with Ockelbo disease in Sweden. American Journal of Tropical Medicine and Hygiene, 33, pp. 1212-1217. Niklasson, B. (1989) Sindbis and Sindbis-like viruses. The Arboviruses: Epideimiology and Ecology, 4, pp. 167-176. Nikolaeva, N.V. (1996) Resurgence of malaria in the former Soviet Union (FSU). Vector Ecology Newsletter, 27, pp. 10-11. Nudelman, S., Galun, R., Kitron, U & Spielman, A. (1988) Physiological characteristics of Culex pipiens populations in the Middle East. Medical and Veterinary Entomology, 2, pp. 161-169. Nuttall, G.H.F., Shipley, A.E., Cobbett, L. & Strangeways-Pigg, T. (1901) The geographical distribution of Anopheles in relation to the former distribution of ague in England. Journal of Hygiene, 1, pp. 4-94. Nye, E.R. (1954) Records of mosquito larvae from Epping Forest. Entomologist’s Monthly Magazine, 90, pp. 119-120. Nye, E.R. (1955) The Flies of the London Area. 11. Culicidae, subfamily Culicinae (Mosquitoes) with a key to the species of Culex. London Naturalist, 34, pp. 114-126. O’Meara, G.F. & Krasnick, G.J. (1970) Dietary and genetic control of the expression of autogenous reproduction in Aedes atropalpus. Journal of Medical Entomology, 7, pp. 328-334. Olejnicek, J. & Gelbic, I. (2000) Autogeny in Culex pipiens quinquefasciatus Say. Journal of Vector Ecology, 25, pp. 118-122. Oshagi, M.A., Chavshin, A.R. & Vatandoost, H. (2006) Analysis of mosquito blood meals using RFLP markers. Experimental Parasitology 114: 259-264. Pant, C.P., Houba, V. & Engers, H.D. (1987) Blood meal identification in vectors. Parasitology Today, 3, pp. 324-326. Patrican, L.A., Hackett, L.E., Briggs, J.E., McGowan, J.W., Unnasch, T.R. & Lee, J.-H. (2007) Host-feeding patterns of Culex mosquitoes in relation to trap habitat. Emerging Infectious Diseases, 13, pp. 1921-1923 Perkins, S.L. & Schall, J.J. (2002) A molecular phylogeny of malarial parasites recovered from Cytochrome b gene sequences. Journal of Parasitology, 88, pp. 972-978.

170

Perez-Tris, J. & Bensch, S. (2005) Dispersal increases local transmission of avian malarial parasites. Ecology Letters, 8, pp. 838-845. Petersen, L.R., Marfin, A.A. & Gubler, D.J. (2003) West Nile Virus. Journal of the American Mosquito Association, 290, pp. 524-528. Peterson, T.A., Komar, N., Komar, O., Navarro-Siguenza, A., Robbins, M.B. & Martinez- Meyer, E. (2004) West Nile virus in the New World: potential impact on bird species. Bird Conservation International, 14, pp. 215-232. Phillips, A., Sabatini, A., Milligan, P.J.M., Boccolini, D., Broomfield, G. & Molyneux, D.H. (1990) The Anopheles maculipennis complex (Diptera: Culicidae): comparison of the cuticular hydrocarbon profiles determined in adults of five Palaearctic species. Bulletin of Entomological Research, 80, pp. 459-464. Pichot, J. & Deruaz, D. (1981) Les Anopheles du complex maculipennis dans la region Lyonnaise. Lyon Mèdyi, 245, pp. 117-121. Pilaski, J. & Mackenstein, H. (1985) Nachweis des Tahyna-Virus bei Steckmüken in zwei vershiedenen europäischen Naturherden. Zentralblatt für Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene B, 180, pp. 394-420. Pilaski, J. (1987) Contributions to the ecology of Tahyna virus in central Europe. Bulletin of the Society for Vector Ecology, 12, pp. 544-553 Platonov, A.E., Shipulin, G.A., Shipulina, O.Y., Tyutyunnik, E.N., Frolochkina, T.I., Lanciotti, R.S., Yazyshina, S., Platonova, O.V., Obukhov, I.L., Zhukov, A.N., Vengerov, Y.Y. & Pokrovskii, V.I. (2001) Outbreak of West Nile virus infection, Volgograd region, Russia, 1999. Emerging Infectious Diseases, 7, pp. 128- 132. Port, G.R. & Boreham, P.F.L. (1980) The relationship of host size to feeding by mosquitoes of the Anopheles gambiae Giles complex (Diptera: Culicaidae). Bulletin of Entomological Research, 70, pp. 133-144. Prina, E., Roux, E., Mattei, D. & Millon, G. (2007) Leishmania DNA is rapidly degraded following parasite death: an analysis by microscopy and real-time PCR. Microbes and Infection, 9, pp. 1307-1315. Proft, J., Maier, W.A. & Kampen, H. (1999) Identification of six sibling species of the Anopheles maculipennis complex (Diptera: Culicidae) by a polymerase chain reaction assay. Parasitology Research, 85, pp. 837-843. Purse, B.V., Mellor, P.S., Rogers, D.J., Samuel, A.R., Mertens, P.P.C. & Baylis, M. (2005) Climate change and the recent emergence of bluetongue in Europe. Nature Reviews Microbiology, 3, pp. 171- 181 171

Ramsdale, C.D. & Gunn, N. (2005) History of and prospects for mosquito-borne disease in Britain. European Mosquito Bulletin, 20, pp. 15-30. Ramsdale, C.D. & Snow, K.R. (1995) Mosquito control in Britain. University of East London. 100pp. Ramsdale, C. & Snow, K. (2000) Distribution of the genus Anopheles in Europe. European Mosquito Bulletin, 7, pp. 1-26. Rappole, J.H., Derrickson, S.R. & Hubalek, Z. (2000) Migratory birds and spread of West Nile virus in the western Hemisphere. Emerging Infectious Diseases, 6, pp. 319-328. Rees, A.T. & Rees, A.E. (1989) Anopheles algeriensis on Anglesey: the story so far. British Mosquito Group Newsletter, 6, pp. 1-5. Rees, A.T. & Snow, K.R. (1990) The distribution of the Genus Anopheles in Britain. Dipterists Digest, 6, pp. 7-19. Rees, A.T. & Snow, K.R. (1992) The distribution of the genus Culex in Britain. Dipterists Digest, 11, pp. 22-32. Rees, A.T. & Snow, K.R. (1994) The distribution of the Genera Coquillettidia, Culiseta and Orthopodomyia in Britain. Dipterists Digest, 1, pp. 36-50. Rees, A.T. & Snow, K.R. (1995) The distribution of the Genus Aedes: subgenera Aedes, Aediomorphus and Finlaya in Britain. Dipterists Digest, 2, pp. 41-48. Rees, A.T. & Snow, K.R. (1996) The distribution Aedes: subgenus Ochlerotatus in Britain. Dipterists Digest, 3, pp. 5-23. Reinert, J.F. (2000) New classification for the composite Genus Aedes (Diptera: Culicidae: Aedini), elevation of subgenus Ochlerotatus to generic rang, reclassification of the other subgenera, and notes on certain subgenera and species. Journal of the American Mosquito Control Association, 16, pp. 175-188. Reinert, J.F. & Harbach R.E. (2005) Generic changes affecting European Aedine mosquitoes (Diptera: Culicidae: Aedini) with a checklist of species. European Mosquito Bulletin, 19, pp. 1-4. Reinert, J.F., Harbach, R.E. & Kitching, I.J. (2004) Phylogeny and classification of Aedini (Diptera: Culicidae), based on morphological characters of all life stages. Zoological Journal of the Linnean Society, 142, pp. 289-368. Reinert, J.F., Harbach, R.E. & Kitching, I.J. (2006) Phylogeny and classification of Finlaya and allied taxa (Diptera: Culicidae: Aedini), based on morphological characters of all life stages. Zoological Journal of the Linnean Society, 148, pp. 1-101.

172

Reinert, J.F., Harbach, R.E. & Kitching, I.J. (2008) Phylogeny and classification of Ochlerotatus and allied taxa (Diptera: Culicidae: Aedini) based on morphological data from all life stages. Zoological Journal of the Linnean Society, 153, pp. 29-114. Reiter, P. (2000) From Shakespeare to Defoe: Malaria in England in the little ice age. Emerging Infectious Diseases, 6, pp. 1-11. Remple, J.D. (2004) Intracellular hematozoa of Raptors: a review and update. Journal of Avian Medicine and Surgery, 18, pp. 75-88. Rezza, G., Nicoletti, L., Angelini, R., Romi, R., Finarelli, A.C., Panning, M., Cordioli, P., Fortuna, C., Boros, S., Magurano, F., Silvi, G., Angelini, P., Dottori, M., Ciufolini, M.G., Majori, G.C. & Cassone, A. (2007) Infection with Chikungunya virus in Italy: an outbreak in a temperate region. The Lancet, 370, pp. 1840-1846. Ribeiro, H., Ramos, H.C., Pires, C.A. & Capela, R.A. (1988) An annotated checklist of the mosquitoes of continental Portugal (Diptera, Culicidae). Actas III Congreso Ibérico de Entomologia, 3, pp. 233-254. Ribeiro, H., Batista, J.L., Ramos, H.C., Pires, C.A., Champalimaud, J.L., Costa, J.M., Araújo, C., Mansinho, K. & Pina, M.C. (1989) An attempt to infect Anopheles atroparvus from Portugal with African Plasmodium falciparum. Revista Portuguesa de Doenças Infecciosas, 12, pp. 81-82. Ricklefs, R.E. & Fallon, S.M. (2002) Diversification and host switching in avian malaria parasites. Proceedings of the Royal Society of London Series B, 269, pp. 885-892. Rodhain, J. & van Hoof, M.T. (1942) Recherches sur l’anophelisme en Belgique. Annals of the Belgium Society of Tropical Medicine, 22, pp. 19-43. Romeo Viamonte, J.M. (1950) Los anofelinos de Espana y de la zona espanola de Marruecos. Revista de sanidad e higiene pública, 24, pp. 213-295. Romi, R., Pierdominici, G., Severini, C., Tamburro, A., Cocchi, M. & Menichetti, D. (1997) Status of malaria vectors in Italy. Journal of medical Entomology, 34, pp. 263-271. Romi, R. (1999) Anopheles labranchiae, an important malaria vector in Italy, and other potential malaria vectors in Southern Europe. European Mosquito Bulletin, 4, pp. 8-10. Romi, R., Boccolini, D., Di Luca, M., La Rosa, G. & Marinucci, M. (2000). Identification of the sibling species of the Anopheles maculipennis complex by heteroduplex analysis. Insect Molecular Biology, 9, pp. 509-513. Romi, R., Sabatinelli, G., & Majori, G. (2001) Could malaria reappear in Italy? Emerging Infectious Diseases, 7, pp. 915-919.

173

Romi, R., Boccolini, D., Hovanesyan, I., Grigoryan, G., Di Luca, M. & Sabatinelli, G. (2002) Anopheles sacharovi (Diptera: Culicidae): A reemerging malaria vector in the Ararat Valley of Armenia. Journal of Medical Entomology, 39, pp. 446-450. Romi, R., Pontuale, M., Ciufolini, G., Fiorentini, G., Marchi, A., Nicoletti, L., Cocchi, M & Tamburro, A. (2004) Potential vectors of West Nile virus following and equine disease outbreak in Italy. Medical and Veterinary Entomology, 18, pp. 14-19. Romoser, W.S., Edman, J.D., Lorenz, L.H. & Scott, T.W. (1989) Histological parameters useful in the identification of multiple blood meals in mosquitoes. American Journal of Tropical Medicine and Hygiene, 41, pp. 737-742. Ross, R. (1897) On some peculiar pigmented cells found intwo mosquitoes fed on malarial blood. British Medical Journal, 2, pp. 1786-1788. Roubaud, E. (1933) Essai synthetique su la vie du mostique commun Culex pipiens. Annales des Sciences Naturelles-Zoologie Et Biologie Animalei, 16, pp. 5-168. Rozas, J., Sanchez-DelBarrio, J.C., Xavier, M. & Rozas, R. (2003) DnaSP, DNA Polymorphism analyses by the coalescent and other methods. Bioinformatics, 19, pp. 2496-2497. Sallum, M.A.M., Schultz, T.R. & Wilkerson, R.C. (2000) Phylogeny of Anophelinae (Diptera: Culicidae) based on morphological characters. Annals of the Entomological Society of America, 93, pp. 745-775. Sallum, M.A.M., Schultz, T.R., Foster, P.G., Aronstein, K., Wirtz, R.A. & Wilkerson, R.C. (2002) Phylogeny of Anophelinae (Diptera: Culicidae) based on nuclear ribosomal and mitochondrial DNA sequences. Systematic Entomology, 27, pp. 361-382. Sartori, M., La Terra, G., Pozzoli, G. & Franzini, C. (1989) A case of autochthonous malaria in Italy. Scandanavian Journal of Infectious Diseases, 21, pp. 357-358. Savage, H.M., Ceinau, C., Nicolescu G., Karabatsos, N., Lanciotti, R., Vladimirescu, A., Laiv, L., Ungureanu, A., Romanca, C. & Tsai, T.F. (1999) Entomologic and avian investigations of an epidemic of West Nile fever in Romania in 1996, with serologic and molecular characterization of a virus isolate from mosquitoes. American Journal of Tropical Medicine and Hygiene, 61, pp. 600-611. Schrenzel, M.D., Maalouf, G.A., Keener, L.L. & Gaffney, P.M. (2003) Molecular characterization of malarial parasites in captive Passerine birds. Journal of Parasitology, 89, pp. 1025-1033. Schultz, J.H., Meier, P.G. & Newson, H.D. (1986) Evolutionary Relationships among the salt marsh Aedes (Diptera: Culicidae). Mosquito Systematics, 18, pp. 145- 195 174

Scott, J.A., Brogdon, W.G & Collins, F.H. (1993). Identification of single specimens of the Anopheles gambiae complex by the polymerase chain-reaction. American Journal of Tropical Medicine and Hygiene, 49, pp. 520-529. Sedaghat, M.M., Linton, Y.-M., Nicolescu, G., Smith, L., Koliopoulos, G., Zounos, A.K., Oshaghi, M.A., Vatandoost, H. & Harbach, R.E. (2003a) Morphological and molecular characterization of Anopheles (Anopheles) sacharovi Favre, a primary vector of malaria in the Middle East. Systematic Entomology, 28, pp. 241–256. Sedaghat, M.M., Linton, Y.-M., Oshaghi, M.A., Vatandoost, H. & Harbach, R.E. (2003b) The Anopheles maculipennis complex (Diptera: Culicidae) in Iran: molecular characterization and recognition of a new species. Bulletin of Entomological Research, 93, pp. 527-535. Service, M.W. (1968a) The taxonomy & biology of two sympatric sibling species of Culex C. pipiens & C. torrentium. Journal of Zoology, 156, pp. 313-323. Service, M.W. (1968b) Observations on feeding and oviposition in some British mosquitoes. Entomologia Experimentalis, 11, pp. 277-285. Service, M. W. (1969) Observations on the ecology of some British mosquitoes. Bulletin of Entomological Research, 59, pp. 161-194. Service, M.W. (1971) Feeding behaviour and host preference of British mosquitoes. Bulletin of Entomological Research, 60, pp. 653-661. Service, M.W. (1993) Mosquito Ecology Field Sampling Methods. Second Edition. Elsevier Science Publication. London pp. 210-290. Shaikevich, E.V. (2007) PCR-RFLP of the COI gene reliably differentiates Cx. pipiens, Cx. pipiens f. molestus and Cx. torrentium of the Pipiens Complex. European Mosquito Bulletin, 23, pp. 25-30. Shepard, J.J., Andreadis, T.G. & Vossbrinck, C.R. (2006) Molecular phylogeny and evolutionary relationships among mosquitoes (Diptera: Culicidae) from the northeastern United States based on small subunit ribosomal DNA (18S rDNA) sequences. Journal of Medical Entomology, 43, pp. 443-454. Shute, P.G. (1933) The life-history and habits of British mosquitoes in relation to their control by anti-larval operations. Journal of Tropical Medicine and Hygiene, 36, pp. 83-88. Shute, P.G. (1954) Indigenous P. vivax malaria in London believed to have been transmitted by An. plumbeus. Monthly Bulletin of the Ministry of Health and Public Health Laboratory Service, 13, pp. 48-51.

175

Shute, P.G. (1963) Indigenious malaria in England since the first World War. The Lancet, 7307, pp. 576-578. Sirinivanakarn, S. (1976) Medical Entomology studies- III. A revision of the subgenus Culex in the Oriental Region (Diptera : Culicidae. Contributions of the American Entomological Institute, 12, pp. 1-278. Smith, J.L. & Fonseca, D.M. (2004) Rapid assays for identification of members of the Culex (Culex) pipiens complex, their hybrids and other sibling species (Diptera: Culicidae). American Journal of Tropical Medicine and Hygiene, 70, pp. 339-345. Smith, J.L., Keyghobadi, N., Matrone, M.A., Escher, R.L. & Fonseca, D.M. (2005) Cross- species comparison of microsatellite loci in the Culex pipiens complex and beyond. Molecular Ecology Notes, 5, pp. 697-700. Snow, K.R. (1990) Mosquitoes. Naturalists’ Handbook 14. Richmond Publishing, England. 66pp. Snow, K. (1998) Distribution of Anopheles mosquitoes in the British Isles. European Mosquito Bulletin, 1, pp. 9-13. Snow, K. (1999) Malaria and Mosquitoes in Britain: the effect of global climate change. European Mosquito Bulletin, 4, pp. 17-25. Snow, K.R., Rees, A.T & Bullbeck, S.J. (1998) A provisional atlas of the mosquitoes of Britain. University of East London, London. 50pp. Snow, K. & Ramsdale, C. (1999) Distribution chart for European mosquitoes European Mosquito Bulletin, 3, pp.14-31. Snow, K. & Medlock, J. (2006) The potential impact of climate change on the distribution of mosquitoes in Britain. European Mosquito Bulletin, 21, pp. 1-10. Snow, K & Medlock, J.M. (2008) The mosquitoes of Epping Forest, Essex, UK. European Mosquito Bulletin, 26, pp. 9-17. Snounou G., Viriyakosol S., Zhu X.P. Jarra, W., Pinheiro, L., do Rosario, V.E., Thaithong S. & Brown, K.N. (1993) High sensitivity of detection of human malaria parasites by the use of nested polymerase chain reaction. Molecular and Biochemical Parasitology, 61, pp. 315-320. Somboon, P., Morakote, N., Koottathep, S. & Trisanarom, U. (1993) Detection of sporozoites of Plasmodium vivax and Plasmodium falciparum in mosquitoes by ELISA: false positivity associated with bovine and swine blood. Transactions of the Royal Society of Tropical Medicine and Hygiene, 87, pp. 322-324.

176

Spanakos, G., Samanidou, A., Tegos, N., Vakalis, N. (2006) Molecular characterisation of common Greek mosquito species. Abstract booklet. 15th European Society of Vector Ecology (SOVE) Meeting, Serres, Greece, 10-14 April 2006. Spielman, A. (1957) The inheritance of autogeny in the Culex pipiens complex of mosquitoes. American Journal of Hygiene, 65, pp. 404-425. Spielman A. (1967) Population structure in the Culex pipiens complex of mosquitoes. Bulletin of the World Health Organization, 37, pp. 271-276. Spielman, A. (1986) Function of mosquito saliva in delivery of pathogens. Havard School of Public Health, Boston Massachusetts, 72pp Spielman, A. (1994) A commentary on research needs for monitoring and containing emergent vector-borne infections. Annals of the New York Academy of Sciences, 740, pp. 457- 461. Spielman, A., (2001) Structure and seasonality of nearctic Culex pipiens population. Annals of the New York Academy of Sciences, 951, pp. 220-234. Staley, J. (1940) A species of mosquito (Diptera, Culicidae) new to Britain. Nature London, 146, pp. 368. Stegnii, V.N. (1976) Revealing of chromosome races in malarial mosquitoes Anopheles sacharovi (Diptera: Culicidae). Tsitologiya, 18, pp. 1039-1041. Stegnii, V.N. & Kabanova, V.M. (1976) Cytological study of indigenous populations of the malaria mosquito in the territory of the U.S.S.R. I. Identification of a new species of Anopheles in the maculipennis complex by the cytodiagnostic method. Meditsinskaya Parazitologiya I Partazitarnye Bolezni 45: 192-198 (in Russian, English translation 1978, Mosquito Systematics, 10, pp. 1-12). Stone, A. (1956) Corrections in the taxonomy and nomenclature of mosquitoes. Proceedings of the Entomological Society of Washington, 58, pp. 333-344. Stone, A., Knight, K.L. & Starcke, H. (1959) A synoptic catalog of the mosquitoes of the world (Diptera: Culicidae). Thomas Say Foundation, 358pp. Su, T. & Mulla, M.S. (1997) Physiological aspects of autogeny in Culex tarsalis (Diptera: Culicidae): influences of sugar-feeding, mating, body weight and wing length. Journal of vector Ecology, 22, pp. 115-121. Suzzoni-Blatger, J. & Sevin, A. (1981) Étude de la chétotaxie larvaire du complexe maculipennis (Diptera: Culicidae) dans la région Toulousaine. Annales de Parasitologie, 57, pp. 640-654.

177

Suzzoni-Blatger, J., Cianchi, R., Bullini, L. & Coluzzi, M. (1990) Le complexe maculipennis: critères morphologiques et enzymatiques de détermination. Annals de Parasitologie Humaine et Comparée, 65, pp. 37-40. Swofford, D. L. (2002). PAUP* Phylogenetic Analysis using Parsimony. Version 4.0b. Sinauer Associates, Sunderland, Massachusetts Takai, K. & Kanda, T. (1986) Phylogenetic relationships among the Anopheles hyrcanus species group based on the degree of hybrid development and comparison with phylogenies by other methods. Journal of Genetics, 61, pp. 235-314. Takken, W. & Kline, D.L. (1989) Carbon dioxide and 1-Octen-3-Ol as mosquito attractants. Journal of the American Mosquito Control Association, 5, pp. 311-316. Takumi, K., Scholte, E-J., Braks, M., Reusken, C., Avenell, D. & Medlock, J. (2008) Introdcution, establishment and seasonal activity of Aedes albopictus in the Netherlands. Abstract booklet 16th Euopean Society for Vector Ecology Conference, Cambridge 25-28 March 2008. Tamura, K., Dudley, J., Nei, M., & Kumar, S. (2007) Molecular Evolutionary Genetic Analyses MEGA software version 4.0. Molecular Biology And Evolution, 24, pp. 1596-1599. Tate, P. & Vincent, M. (1936) The biology of autogenous and anautogenous races of Culex pipiens L. Parasitology, 28, pp. 115-145. Tavernier, P., Sagesse, M., Van Wettere, A. & Redig, P. (2005) Malaria in an Eastern Screech owl (Otus asio). Avian Diseases, 49, pp. 433-435. Taylor, D. (1988) Hybridization of Aedes (Protomacleaya) zoosophus with Ae. (Pro.) triseriatus group species: hybrid morphology. Journal of the American Mosquito Control Association, 4, pp. 23-28. Tenorio, J. (1977) Revision of the genus Bironella (Diptera: Culicidae). Journal of Medical Entomology, 14, pp. 317-361. Theobald, F.V (1910) A monograph of the Culicidae of the World. London, England 646pp. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. & Higgins, D.G. (1997) The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research, 24, pp. 4876-4882. Tsai, T. F., Popovici, F., Cernescu, C., Campbell, G. L., Nedelcu, N. I. (1998) West Nile encephalitis epidemic in southeastern Romania. Lancet, 352, pp. 767-771.

178

Turell, M.J., Presley, S.M., Gad, A.M., Cope, S.E., Dohm, D.J., Morrill, J.C. & Arthur, R.R. (1996) Vector competence of Egyptian mosquitoes for Rift Valley fever virus. American Journal of Tropical Medicine and Hygiene, 54, pp. 136-139. Tveten, M.S. & Meola, R.W. (1988) Autogeny in Culex salinarius from Texas, Florida and New Jersey. Journal of the American Mosquito Control Association, 4, pp. 436-441. Tyagi, B.K (2003) A handbook of medically important insects and other arthropods. India, Scientific Publishers, pp 47. Ungureanu, E. & Shute, P.G. (1947) The value of the wing scales as an aid to the taxonomy of adult Anopheles maculipennis. The Proceedings of the Royal Entomological Society of London, Series B, Taxonomy, 16, pp. 79-85. Urbanelli, S., Silvestrini, F., Reisen, W.K., De Vito, E. & Bullini, L. (1997) Californian hybrid zone between Culex pipiens pipiens and Culex p. quinquefasciatus revisted (Diptera: Culicidae). Journal of Medical Entomology, 34, pp. 116-127. Van der Meulen, K.M., Pensaert, M.B. & Nauwynck, H.J. (2005) West Nile virus in the vertebrate world. Archives in Virology, 150, pp. 637-657. Van Den Hurk, A.F., Smith, I.L. & Smith, G.A. (2007) Development and evaluation of real- time polymerase chain reaction assays to identify mosquito (Diptera: Culicidae) blood meals originating from native Australian mammals. Journal of Medical Entomology, 44, pp. 85-92. van Riper, C., III, Atkinson, C.T. & Seed T.M. (1993) Plasmodia of birds. In: J.P. Kreier, ed. Parasitic protozo, Vol 4, 2nd edition. Academic Press, San Diego. pp. 73-140. van Thiel, P.H. (1927) Sur l'origine des variations de taille de l'Anopheles maculipennis dans les Pays-Bas. Bulletin de la Société de Pathologie Exototique, 20, pp. 66-390. Vinogradova, Y.B. (1966) Blood-sucking mosquitoes of the Culex pipiens L. Complex (Diptera: Culicidae), their practical importance, taxonomy and biology. Entomological Review, 41, pp 1-4. Vinogradova, Y.E.B. & Fomenko, R.B. (1968) Morphology and Biology of Culex pipiens L. (Diptera: Culicidae) in Uzbekistan. Entomology Reviews, 47, pp. 1-4. Vinogradova, E.B., Shaikevich, E.V. & Ivanitsky, A.V. (2007) A study of the distribution of the Culex pipiens complex (Insecta: Diptera: Culicidae) mosquitoes in the European part of Russia by molecular methods of identification. Comparative Cytogenetics, 1, pp.129-138. Vladimirescu, A.L., Nicolescu, G., Coipan C.E., Ciulacu-Purcarea, V. Ionescu, L., Bicheru, S., Stefanoiu, V. & Steriu, D. (2006) The presence of Anopheles daciae (Diptera: 179

Culicidae) in close proximity of malaria indigenous case in Romania. Abstract booklet of the 15th European Society for Vector Ecology (ESOVE) meeting, Serres, Greece, 10-14 April 2006. Vogler, A.P. & DeSalle, R. (1994) Evolution and phylogenetic information content of the ITS- 1 region in the Tiger Beetle Cicindela dorsalis. Molecular Biology and Evolution, 11, pp. 393-405. Wallace, F.E. (1958) British mosquitoes and their control. Economic Series 4A, 4th Edition, British Museum (Natural History), London. 25pp. Weitz, B. (1956) Identification of blood meals of blood-sucking arthropods. Bulletin of the World Health Organization, 15, pp. 473-490. Wesson, D. M., McLain, D. K., Oliver, J.H., Piesman, J. & Collins, F.H. (1993) Investigation of the validity of species status of Ixodes dammini (Acari: Ixodidae) using rDNA. Proceedings of the National Academy of Sciences U.S.A., 90, pp. 10221-10225. Weyer, F. (1942) Bestimmungsschluessel Fuer die Anopheles-Weibchen und- Larven in Europa, Nordafrika und Westasien. Deutsche Tropenmedizinicshe Zeitschrift, 46, pp. 1-35. White, G.B. (1978) Systematic reappraisal of the Anopheles maculipennis complex. Mosquito Systematics, 10, pp. 3-44. White, G.B. (1985) Anopheles bwambae sp.n., a malaria vector in the Semliki Valley, Uganda, and its relationships with other sibling species of the An. gambiae complex (Diptera: Culicidae). Systematic Entomology, 10, pp. 501-522. Whitfield, D., Curtis, C.F., White, G.B., Targett, G.A.T., Warhurst, D.C. & Bradley, D.J. (1984) Two cases of falciparum malaria acquired in Britain. British Medical Journal, 289, pp. 1607-1609. Wikelski, M., Foufopoulos, J., Vargas, H. & Snell, H. (2004) Galapagos birds and diseases: invasive pathogens as threats for island species. Ecology and Society, 9, pp. 5-14. Wilkerson, R.C., Foster, P.G., Li, Cong & Sallum, M.A.M . (2005) Molecular phylogeny of neotropical Anopheles (Nyssorhynchus) albitarsis species complex (Diptera: Culicidae). Annals of the Entomological Society of America, 98, pp. 918-925. Wood, M.J. & Cosgrove, C.L. (2006) The hitchhiker’s guide to avian malaria. TRENDS in Ecology and Evolution, 21, pp. 5-7. Wright, W.R. (1924) The mosquitoes of North Wales. Entomologist’s Monthly Magazine, 60, pp. 208-210.

180

Walter Reed Biosystematics Unit (WRBU) (2008) Online Systematic Catalogue of Culicidae. Available online at http://www.wrbu.org/index.html (accessed: 31 January 2009) Zavortink, T.J. (1990) Classical taxonomy of mosquitoes- a memorial to John N. Belkin. Journal of the American Mosquito Control Association, 6, pp. 593-599. Zimmerman, R.H., Galardo, A.K.R., Lounibos, L.P., Arruda, M. & Wirtz, R. (2006) Bloodmeal hosts of Anopheles species (Diptera: Culicidae) in a malaria-endemic area of the Brazilian Amazon. Journal of Medical Entomology, 43, pp 947-956.

181